[Dynamo] initial foreach_op impl

.ci/aarch64_linux/aarch64_ci_build.sh

@@ -3,15 +3,22 @@ set -eux -o pipefail
 
 GPU_ARCH_VERSION=${GPU_ARCH_VERSION:-}
 
-if [[ "$GPU_ARCH_VERSION" == *"12.6"* ]]; then
-    export TORCH_CUDA_ARCH_LIST="9.0"
-elif [[ "$GPU_ARCH_VERSION" == *"12.8"* ]]; then
-    export TORCH_CUDA_ARCH_LIST="9.0;10.0;12.0"
-fi
-
 SCRIPTPATH="$( cd -- "$(dirname "$0")" >/dev/null 2>&1 ; pwd -P )"
 source $SCRIPTPATH/aarch64_ci_setup.sh
 
+tagged_version() {
+  GIT_DESCRIBE="git --git-dir /pytorch/.git describe --tags --match v[0-9]*.[0-9]*.[0-9]*"
+  if ${GIT_DESCRIBE} --exact >/dev/null; then
+    ${GIT_DESCRIBE}
+  else
+    return 1
+  fi
+}
+
+if tagged_version >/dev/null; then
+  export OVERRIDE_PACKAGE_VERSION="$(tagged_version | sed -e 's/^v//' -e 's/-.*$//')"
+fi
+
 ###############################################################################
 # Run aarch64 builder python
 ###############################################################################

.ci/aarch64_linux/aarch64_ci_setup.sh

@@ -5,14 +5,16 @@ set -eux -o pipefail
 # By creating symlinks from desired /opt/python to /usr/local/bin/
 
 NUMPY_VERSION=2.0.2
-if [[ "$DESIRED_PYTHON"  == "3.13" || "$DESIRED_PYTHON" == "3.13t" ]]; then
+PYGIT2_VERSION=1.15.1
+if [[ "$DESIRED_PYTHON"  == "3.13" ]]; then
     NUMPY_VERSION=2.1.2
+    PYGIT2_VERSION=1.16.0
 fi
 
 SCRIPTPATH="$( cd "$(dirname "$0")" ; pwd -P )"
 source $SCRIPTPATH/../manywheel/set_desired_python.sh
 
-pip install -q numpy==${NUMPY_VERSION} pyyaml==6.0.2 scons==4.7.0 ninja==1.11.1 patchelf==0.17.2
+pip install -q numpy==${NUMPY_VERSION} pyyaml==6.0.2 scons==4.7.0 ninja==1.11.1 patchelf==0.17.2 pygit2==${PYGIT2_VERSION}
 
 for tool in python python3 pip pip3 ninja scons patchelf; do
     ln -sf ${DESIRED_PYTHON_BIN_DIR}/${tool} /usr/local/bin;

.ci/aarch64_linux/aarch64_wheel_ci_build.py

@@ -4,9 +4,12 @@
 import os
 import shutil
 from subprocess import check_call, check_output
+from typing import List
+
+from pygit2 import Repository
 
 
-def list_dir(path: str) -> list[str]:
+def list_dir(path: str) -> List[str]:
     """'
     Helper for getting paths for Python
     """
@@ -55,7 +58,7 @@ def build_ArmComputeLibrary() -> None:
         shutil.copytree(f"{acl_checkout_dir}/{d}", f"{acl_install_dir}/{d}")
 
 
-def update_wheel(wheel_path, desired_cuda) -> None:
+def update_wheel(wheel_path) -> None:
     """
     Update the cuda wheel libraries
     """
@@ -77,6 +80,7 @@ def update_wheel(wheel_path, desired_cuda) -> None:
         "/usr/local/cuda/lib64/libnvToolsExt.so.1",
         "/usr/local/cuda/lib64/libnvJitLink.so.12",
         "/usr/local/cuda/lib64/libnvrtc.so.12",
+        "/usr/local/cuda/lib64/libnvrtc-builtins.so.12.4",
         "/usr/local/cuda/lib64/libcudnn_adv.so.9",
         "/usr/local/cuda/lib64/libcudnn_cnn.so.9",
         "/usr/local/cuda/lib64/libcudnn_graph.so.9",
@@ -96,14 +100,6 @@ def update_wheel(wheel_path, desired_cuda) -> None:
             "/usr/local/lib/libnvpl_lapack_core.so.0",
             "/usr/local/lib/libnvpl_blas_core.so.0",
         ]
-        if "126" in desired_cuda:
-            libs_to_copy += [
-                "/usr/local/cuda/lib64/libnvrtc-builtins.so.12.6",
-            ]
-        elif "128" in desired_cuda:
-            libs_to_copy += [
-                "/usr/local/cuda/lib64/libnvrtc-builtins.so.12.8",
-            ]
     else:
         libs_to_copy += [
             "/opt/OpenBLAS/lib/libopenblas.so.0",
@@ -175,22 +171,22 @@ if __name__ == "__main__":
     args = parse_arguments()
     enable_mkldnn = args.enable_mkldnn
     enable_cuda = args.enable_cuda
-    branch = check_output(
-        ["git", "rev-parse", "--abbrev-ref", "HEAD"], cwd="/pytorch"
-    ).decode()
+    repo = Repository("/pytorch")
+    branch = repo.head.name
+    if branch == "HEAD":
+        branch = "master"
 
     print("Building PyTorch wheel")
     build_vars = "MAX_JOBS=5 CMAKE_SHARED_LINKER_FLAGS=-Wl,-z,max-page-size=0x10000 "
     os.system("cd /pytorch; python setup.py clean")
 
     override_package_version = os.getenv("OVERRIDE_PACKAGE_VERSION")
-    desired_cuda = os.getenv("DESIRED_CUDA")
     if override_package_version is not None:
         version = override_package_version
         build_vars += (
             f"BUILD_TEST=0 PYTORCH_BUILD_VERSION={version} PYTORCH_BUILD_NUMBER=1 "
         )
-    elif branch in ["nightly", "main"]:
+    elif branch in ["nightly", "master"]:
         build_date = (
             check_output(["git", "log", "--pretty=format:%cs", "-1"], cwd="/pytorch")
             .decode()
@@ -200,6 +196,7 @@ if __name__ == "__main__":
             check_output(["cat", "version.txt"], cwd="/pytorch").decode().strip()[:-2]
         )
         if enable_cuda:
+            desired_cuda = os.getenv("DESIRED_CUDA")
             build_vars += f"BUILD_TEST=0 PYTORCH_BUILD_VERSION={version}.dev{build_date}+{desired_cuda} PYTORCH_BUILD_NUMBER=1 "
         else:
             build_vars += f"BUILD_TEST=0 PYTORCH_BUILD_VERSION={version}.dev{build_date} PYTORCH_BUILD_NUMBER=1 "
@@ -228,6 +225,6 @@ if __name__ == "__main__":
         print("Updating Cuda Dependency")
         filename = os.listdir("/pytorch/dist/")
         wheel_path = f"/pytorch/dist/{filename[0]}"
-        update_wheel(wheel_path, desired_cuda)
+        update_wheel(wheel_path)
     pytorch_wheel_name = complete_wheel("/pytorch/")
     print(f"Build Complete. Created {pytorch_wheel_name}..")

.ci/aarch64_linux/build_aarch64_wheel.py

@@ -12,7 +12,7 @@ import os
 import subprocess
 import sys
 import time
-from typing import Optional, Union
+from typing import Dict, List, Optional, Tuple, Union
 
 import boto3
 
@@ -24,12 +24,10 @@ os_amis = {
     "ubuntu22_04": "ami-0c6c29c5125214c77",  # login_name: ubuntu
     "redhat8": "ami-0698b90665a2ddcf1",  # login_name: ec2-user
 }
-
 ubuntu18_04_ami = os_amis["ubuntu18_04"]
-ubuntu20_04_ami = os_amis["ubuntu20_04"]
 
 
-def compute_keyfile_path(key_name: Optional[str] = None) -> tuple[str, str]:
+def compute_keyfile_path(key_name: Optional[str] = None) -> Tuple[str, str]:
     if key_name is None:
         key_name = os.getenv("AWS_KEY_NAME")
         if key_name is None:
@@ -59,7 +57,7 @@ def ec2_instances_by_id(instance_id):
 
 
 def start_instance(
-    key_name, ami=ubuntu20_04_ami, instance_type="t4g.2xlarge", ebs_size: int = 50
+    key_name, ami=ubuntu18_04_ami, instance_type="t4g.2xlarge", ebs_size: int = 50
 ):
     inst = ec2.create_instances(
         ImageId=ami,
@@ -98,7 +96,7 @@ class RemoteHost:
         self.keyfile_path = keyfile_path
         self.login_name = login_name
 
-    def _gen_ssh_prefix(self) -> list[str]:
+    def _gen_ssh_prefix(self) -> List[str]:
         return [
             "ssh",
             "-o",
@@ -110,13 +108,13 @@ class RemoteHost:
         ]
 
     @staticmethod
-    def _split_cmd(args: Union[str, list[str]]) -> list[str]:
+    def _split_cmd(args: Union[str, List[str]]) -> List[str]:
         return args.split() if isinstance(args, str) else args
 
-    def run_ssh_cmd(self, args: Union[str, list[str]]) -> None:
+    def run_ssh_cmd(self, args: Union[str, List[str]]) -> None:
         subprocess.check_call(self._gen_ssh_prefix() + self._split_cmd(args))
 
-    def check_ssh_output(self, args: Union[str, list[str]]) -> str:
+    def check_ssh_output(self, args: Union[str, List[str]]) -> str:
         return subprocess.check_output(
             self._gen_ssh_prefix() + self._split_cmd(args)
         ).decode("utf-8")
@@ -159,7 +157,7 @@ class RemoteHost:
     def using_docker(self) -> bool:
         return self.container_id is not None
 
-    def run_cmd(self, args: Union[str, list[str]]) -> None:
+    def run_cmd(self, args: Union[str, List[str]]) -> None:
         if not self.using_docker():
             return self.run_ssh_cmd(args)
         assert self.container_id is not None
@@ -180,7 +178,7 @@ class RemoteHost:
         if rc != 0:
             raise subprocess.CalledProcessError(rc, docker_cmd)
 
-    def check_output(self, args: Union[str, list[str]]) -> str:
+    def check_output(self, args: Union[str, List[str]]) -> str:
         if not self.using_docker():
             return self.check_ssh_output(args)
         assert self.container_id is not None
@@ -232,7 +230,7 @@ class RemoteHost:
             )
         self.download_file(remote_file, local_file)
 
-    def list_dir(self, path: str) -> list[str]:
+    def list_dir(self, path: str) -> List[str]:
         return self.check_output(["ls", "-1", path]).split("\n")
 
 
@@ -360,7 +358,7 @@ def checkout_repo(
     branch: str = "main",
     url: str,
     git_clone_flags: str,
-    mapping: dict[str, tuple[str, str]],
+    mapping: Dict[str, Tuple[str, str]],
 ) -> Optional[str]:
     for prefix in mapping:
         if not branch.startswith(prefix):
@@ -621,11 +619,9 @@ def build_torchaudio(
     if host.using_docker():
         build_vars += " CMAKE_SHARED_LINKER_FLAGS=-Wl,-z,max-page-size=0x10000"
 
-    host.run_cmd(
-        f"cd audio && export FFMPEG_ROOT=$(pwd)/third_party/ffmpeg && export USE_FFMPEG=1 \
+    host.run_cmd(f"cd audio && export FFMPEG_ROOT=$(pwd)/third_party/ffmpeg && export USE_FFMPEG=1 \
         && ./packaging/ffmpeg/build.sh \
-        && {build_vars} python3 setup.py bdist_wheel"
-    )
+        && {build_vars} python3 setup.py bdist_wheel")
 
     wheel_name = host.list_dir("audio/dist")[0]
     embed_libgomp(host, use_conda, os.path.join("audio", "dist", wheel_name))
@@ -683,7 +679,7 @@ def build_domains(
     branch: str = "main",
     use_conda: bool = True,
     git_clone_flags: str = "",
-) -> tuple[str, str, str, str]:
+) -> Tuple[str, str, str, str]:
     vision_wheel_name = build_torchvision(
         host, branch=branch, use_conda=use_conda, git_clone_flags=git_clone_flags
     )
@@ -710,7 +706,7 @@ def start_build(
     pytorch_build_number: Optional[str] = None,
     shallow_clone: bool = True,
     enable_mkldnn: bool = False,
-) -> tuple[str, str, str, str, str]:
+) -> Tuple[str, str, str, str, str]:
     git_clone_flags = " --depth 1 --shallow-submodules" if shallow_clone else ""
     if host.using_docker() and not use_conda:
         print("Auto-selecting conda option for docker images")
@@ -934,9 +930,9 @@ def parse_arguments():
     parser.add_argument("--debug", action="store_true")
     parser.add_argument("--build-only", action="store_true")
     parser.add_argument("--test-only", type=str)
-    group = parser.add_mutually_exclusive_group()
-    group.add_argument("--os", type=str, choices=list(os_amis.keys()))
-    group.add_argument("--ami", type=str)
+    parser.add_argument(
+        "--os", type=str, choices=list(os_amis.keys()), default="ubuntu20_04"
+    )
     parser.add_argument(
         "--python-version",
         type=str,
@@ -966,13 +962,7 @@ def parse_arguments():
 
 if __name__ == "__main__":
     args = parse_arguments()
-    ami = (
-        args.ami
-        if args.ami is not None
-        else os_amis[args.os]
-        if args.os is not None
-        else ubuntu20_04_ami
-    )
+    ami = os_amis[args.os]
     keyfile_path, key_name = compute_keyfile_path(args.key_name)
 
     if args.list_instances:

.ci/docker/aotriton_version.txt

@@ -0,0 +1,5 @@
+0.7b
+manylinux_2_17
+rocm6.2
+9be04068c3c0857a4cfd17d7e39e71d0423ebac2
+3e9e1959d23b93d78a08fcc5f868125dc3854dece32fd9458be9ef4467982291

.ci/docker/build.sh

@@ -86,10 +86,6 @@ CMAKE_VERSION=3.18.5
 
 _UCX_COMMIT=7bb2722ff2187a0cad557ae4a6afa090569f83fb
 _UCC_COMMIT=20eae37090a4ce1b32bcce6144ccad0b49943e0b
-if [[ "$image" == *rocm* ]]; then
-  _UCX_COMMIT=cc312eaa4655c0cc5c2bcd796db938f90563bcf6
-  _UCC_COMMIT=0c0fc21559835044ab107199e334f7157d6a0d3d
-fi
 
 # It's annoying to rename jobs every time you want to rewrite a
 # configuration, so we hardcode everything here rather than do it
@@ -109,6 +105,20 @@ case "$image" in
     CONDA_CMAKE=yes
     TRITON=yes
     ;;
+  pytorch-linux-focal-cuda12.1-cudnn9-py3-gcc9)
+    CUDA_VERSION=12.1.1
+    CUDNN_VERSION=9
+    ANACONDA_PYTHON_VERSION=3.10
+    GCC_VERSION=9
+    PROTOBUF=yes
+    DB=yes
+    VISION=yes
+    KATEX=yes
+    UCX_COMMIT=${_UCX_COMMIT}
+    UCC_COMMIT=${_UCC_COMMIT}
+    CONDA_CMAKE=yes
+    TRITON=yes
+    ;;
   pytorch-linux-focal-cuda12.4-cudnn9-py3-gcc9-inductor-benchmarks)
     CUDA_VERSION=12.4.1
     CUDNN_VERSION=9
@@ -124,6 +134,36 @@ case "$image" in
     TRITON=yes
     INDUCTOR_BENCHMARKS=yes
     ;;
+  pytorch-linux-focal-cuda12.1-cudnn9-py3-gcc9-inductor-benchmarks)
+    CUDA_VERSION=12.1.1
+    CUDNN_VERSION=9
+    ANACONDA_PYTHON_VERSION=3.10
+    GCC_VERSION=9
+    PROTOBUF=yes
+    DB=yes
+    VISION=yes
+    KATEX=yes
+    UCX_COMMIT=${_UCX_COMMIT}
+    UCC_COMMIT=${_UCC_COMMIT}
+    CONDA_CMAKE=yes
+    TRITON=yes
+    INDUCTOR_BENCHMARKS=yes
+    ;;
+  pytorch-linux-focal-cuda12.1-cudnn9-py3.12-gcc9-inductor-benchmarks)
+    CUDA_VERSION=12.1.1
+    CUDNN_VERSION=9
+    ANACONDA_PYTHON_VERSION=3.12
+    GCC_VERSION=9
+    PROTOBUF=yes
+    DB=yes
+    VISION=yes
+    KATEX=yes
+    UCX_COMMIT=${_UCX_COMMIT}
+    UCC_COMMIT=${_UCC_COMMIT}
+    CONDA_CMAKE=yes
+    TRITON=yes
+    INDUCTOR_BENCHMARKS=yes
+    ;;
   pytorch-linux-focal-cuda12.4-cudnn9-py3.12-gcc9-inductor-benchmarks)
     CUDA_VERSION=12.4.1
     CUDNN_VERSION=9
@@ -139,10 +179,10 @@ case "$image" in
     TRITON=yes
     INDUCTOR_BENCHMARKS=yes
     ;;
-  pytorch-linux-focal-cuda12.4-cudnn9-py3.13-gcc9-inductor-benchmarks)
-    CUDA_VERSION=12.4.1
+  pytorch-linux-focal-cuda11.8-cudnn9-py3-gcc9)
+    CUDA_VERSION=11.8.0
     CUDNN_VERSION=9
-    ANACONDA_PYTHON_VERSION=3.13
+    ANACONDA_PYTHON_VERSION=3.10
     GCC_VERSION=9
     PROTOBUF=yes
     DB=yes
@@ -152,10 +192,37 @@ case "$image" in
     UCC_COMMIT=${_UCC_COMMIT}
     CONDA_CMAKE=yes
     TRITON=yes
-    INDUCTOR_BENCHMARKS=yes
     ;;
-  pytorch-linux-focal-cuda11.8-cudnn9-py3-gcc9)
-    CUDA_VERSION=11.8.0
+  pytorch-linux-focal-cuda12.4-cudnn9-py3-gcc9)
+    CUDA_VERSION=12.4.1
+    CUDNN_VERSION=9
+    ANACONDA_PYTHON_VERSION=3.10
+    GCC_VERSION=9
+    PROTOBUF=yes
+    DB=yes
+    VISION=yes
+    KATEX=yes
+    UCX_COMMIT=${_UCX_COMMIT}
+    UCC_COMMIT=${_UCC_COMMIT}
+    CONDA_CMAKE=yes
+    TRITON=yes
+    ;;
+  pytorch-linux-focal-cuda12.1-cudnn9-py3-gcc9)
+    CUDA_VERSION=12.1.1
+    CUDNN_VERSION=9
+    ANACONDA_PYTHON_VERSION=3.10
+    GCC_VERSION=9
+    PROTOBUF=yes
+    DB=yes
+    VISION=yes
+    KATEX=yes
+    UCX_COMMIT=${_UCX_COMMIT}
+    UCC_COMMIT=${_UCC_COMMIT}
+    CONDA_CMAKE=yes
+    TRITON=yes
+    ;;
+  pytorch-linux-focal-cuda12.4-cudnn9-py3-gcc9)
+    CUDA_VERSION=12.4.1
     CUDNN_VERSION=9
     ANACONDA_PYTHON_VERSION=3.10
     GCC_VERSION=9
@@ -210,7 +277,18 @@ case "$image" in
     ;;
   pytorch-linux-focal-rocm-n-1-py3)
     ANACONDA_PYTHON_VERSION=3.10
-    GCC_VERSION=11
+    GCC_VERSION=9
+    PROTOBUF=yes
+    DB=yes
+    VISION=yes
+    ROCM_VERSION=6.1
+    NINJA_VERSION=1.9.0
+    CONDA_CMAKE=yes
+    TRITON=yes
+    ;;
+  pytorch-linux-focal-rocm-n-py3)
+    ANACONDA_PYTHON_VERSION=3.10
+    GCC_VERSION=9
     PROTOBUF=yes
     DB=yes
     VISION=yes
@@ -218,25 +296,6 @@ case "$image" in
     NINJA_VERSION=1.9.0
     CONDA_CMAKE=yes
     TRITON=yes
-    KATEX=yes
-    UCX_COMMIT=${_UCX_COMMIT}
-    UCC_COMMIT=${_UCC_COMMIT}
-    INDUCTOR_BENCHMARKS=yes
-    ;;
-  pytorch-linux-focal-rocm-n-py3)
-    ANACONDA_PYTHON_VERSION=3.10
-    GCC_VERSION=11
-    PROTOBUF=yes
-    DB=yes
-    VISION=yes
-    ROCM_VERSION=6.3
-    NINJA_VERSION=1.9.0
-    CONDA_CMAKE=yes
-    TRITON=yes
-    KATEX=yes
-    UCX_COMMIT=${_UCX_COMMIT}
-    UCC_COMMIT=${_UCC_COMMIT}
-    INDUCTOR_BENCHMARKS=yes
     ;;
   pytorch-linux-jammy-xpu-2024.0-py3)
     ANACONDA_PYTHON_VERSION=3.9
@@ -451,7 +510,7 @@ docker build \
        --build-arg "NINJA_VERSION=${NINJA_VERSION:-}" \
        --build-arg "KATEX=${KATEX:-}" \
        --build-arg "ROCM_VERSION=${ROCM_VERSION:-}" \
-       --build-arg "PYTORCH_ROCM_ARCH=${PYTORCH_ROCM_ARCH:-gfx90a;gfx942}" \
+       --build-arg "PYTORCH_ROCM_ARCH=${PYTORCH_ROCM_ARCH:-gfx90a}" \
        --build-arg "IMAGE_NAME=${IMAGE_NAME}" \
        --build-arg "UCX_COMMIT=${UCX_COMMIT}" \
        --build-arg "UCC_COMMIT=${UCC_COMMIT}" \

.ci/docker/centos-rocm/Dockerfile

@@ -113,6 +113,13 @@ COPY triton_version.txt triton_version.txt
 RUN if [ -n "${TRITON}" ]; then bash ./install_triton.sh; fi
 RUN rm install_triton.sh common_utils.sh triton.txt triton_version.txt
 
+# Install AOTriton (Early fail)
+COPY ./aotriton_version.txt aotriton_version.txt
+COPY ./common/common_utils.sh common_utils.sh
+COPY ./common/install_aotriton.sh install_aotriton.sh
+RUN ["/bin/bash", "-c", "./install_aotriton.sh /opt/rocm && rm -rf install_aotriton.sh aotriton_version.txt common_utils.sh"]
+ENV AOTRITON_INSTALLED_PREFIX /opt/rocm/aotriton
+
 # Install ccache/sccache (do this last, so we get priority in PATH)
 COPY ./common/install_cache.sh install_cache.sh
 ENV PATH /opt/cache/bin:$PATH

.ci/docker/ci_commit_pins/executorch.txt

@@ -1 +1 @@
-5e4d6b6380d575e48e37e9d987fded4ec588e7bc
+6f638937d64e3396793956d75ee3e14802022745

.ci/docker/ci_commit_pins/nccl-cu11.txt

@@ -1 +0,0 @@
-v2.21.5-1

.ci/docker/ci_commit_pins/nccl-cu12.txt

@@ -1 +0,0 @@
-v2.25.1-1

.ci/docker/ci_commit_pins/timm.txt

@@ -1 +1 @@
-5d535d7a2d4b435b1b5c1177fd8f04a12b942b9a
+ac3470188b914c5d7a5058a7e28b9eb685a62427

.ci/docker/ci_commit_pins/triton.txt

@@ -1 +1 @@
-4b3bb1f8da0ded6ccd572dd1358ef45af5a1befe
+35c6c7c6284582b3f41c71c150e11b517acf074a

.ci/docker/common/install_acl.sh

@@ -1,7 +1,7 @@
 set -euo pipefail
 
 readonly version=v24.04
-readonly src_host=https://github.com/ARM-software
+readonly src_host=https://review.mlplatform.org/ml
 readonly src_repo=ComputeLibrary
 
 # Clone ACL

.ci/docker/common/install_aotriton.sh

@@ -0,0 +1,23 @@
+#!/bin/bash
+
+set -ex
+
+source "$(dirname "${BASH_SOURCE[0]}")/common_utils.sh"
+
+TARBALL='aotriton.tar.gz'
+# This read command alwasy returns with exit code 1
+read -d "\n" VER MANYLINUX ROCMBASE PINNED_COMMIT SHA256 < aotriton_version.txt || true
+ARCH=$(uname -m)
+AOTRITON_INSTALL_PREFIX="$1"
+AOTRITON_URL="https://github.com/ROCm/aotriton/releases/download/${VER}/aotriton-${VER}-${MANYLINUX}_${ARCH}-${ROCMBASE}-shared.tar.gz"
+
+cd "${AOTRITON_INSTALL_PREFIX}"
+# Must use -L to follow redirects
+curl -L --retry 3 -o "${TARBALL}" "${AOTRITON_URL}"
+ACTUAL_SHA256=$(sha256sum "${TARBALL}" | cut -d " " -f 1)
+if [ "${SHA256}" != "${ACTUAL_SHA256}" ]; then
+  echo -n "Error: The SHA256 of downloaded tarball is ${ACTUAL_SHA256},"
+  echo " which does not match the expected value ${SHA256}."
+  exit
+fi
+tar xf "${TARBALL}" && rm -rf "${TARBALL}"

.ci/docker/common/install_base.sh

@@ -32,12 +32,8 @@ install_ubuntu() {
 
   # HACK: UCC testing relies on libnccl library from NVIDIA repo, and version 2.16 crashes
   # See https://github.com/pytorch/pytorch/pull/105260#issuecomment-1673399729
-  # TODO: Eliminate this hack, we should not relay on apt-get installation
-  # See https://github.com/pytorch/pytorch/issues/144768
   if [[ "$UBUNTU_VERSION" == "20.04"* && "$CUDA_VERSION" == "11.8"* ]]; then
     maybe_libnccl_dev="libnccl2=2.15.5-1+cuda11.8 libnccl-dev=2.15.5-1+cuda11.8 --allow-downgrades --allow-change-held-packages"
-  elif [[ "$UBUNTU_VERSION" == "20.04"* && "$CUDA_VERSION" == "12.4"* ]]; then
-    maybe_libnccl_dev="libnccl2=2.25.1-1+cuda12.4 libnccl-dev=2.25.1-1+cuda12.4 --allow-downgrades --allow-change-held-packages"
   else
     maybe_libnccl_dev=""
   fi
@@ -80,8 +76,7 @@ install_ubuntu() {
     vim \
     unzip \
     gpg-agent \
-    gdb \
-    bc
+    gdb
 
   # Should resolve issues related to various apt package repository cert issues
   # see: https://github.com/pytorch/pytorch/issues/65931

.ci/docker/common/install_cache.sh

@@ -9,7 +9,12 @@ install_ubuntu() {
   # Instead use lib and headers from OpenSSL1.1 installed in `install_openssl.sh``
   apt-get install -y cargo
   echo "Checking out sccache repo"
-  git clone https://github.com/mozilla/sccache -b v0.9.1
+  if [ -n "$CUDA_VERSION" ]; then
+      # TODO: Remove this
+      git clone https://github.com/pytorch/sccache
+  else
+      git clone https://github.com/mozilla/sccache -b v0.8.2
+  fi
   cd sccache
   echo "Building sccache"
   cargo build --release
@@ -36,33 +41,41 @@ sed -e 's|PATH="\(.*\)"|PATH="/opt/cache/bin:\1"|g' -i /etc/environment
 export PATH="/opt/cache/bin:$PATH"
 
 # Setup compiler cache
-install_ubuntu
+if [ -n "$ROCM_VERSION" ]; then
+  curl --retry 3 http://repo.radeon.com/misc/.sccache_amd/sccache -o /opt/cache/bin/sccache
+else
+  ID=$(grep -oP '(?<=^ID=).+' /etc/os-release | tr -d '"')
+  if [ -n "$CUDA_VERSION" ]; then
+    # TODO: Install the pre-built binary from S3 as building from source
+    # https://github.com/pytorch/sccache has started failing mysteriously
+    # in which sccache server couldn't start with the following error:
+    #   sccache: error: Invalid argument (os error 22)
+    install_binary
+  else
+    install_ubuntu
+  fi
+fi
 chmod a+x /opt/cache/bin/sccache
 
 function write_sccache_stub() {
   # Unset LD_PRELOAD for ps because of asan + ps issues
   # https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90589
   if [ $1 == "gcc" ]; then
-    # Do not call sccache recursively when dumping preprocessor argument
-    # For some reason it's very important for the first cached nvcc invocation
-    cat >"/opt/cache/bin/$1" <<EOF
+  # Do not call sccache recursively when dumping preprocessor argument
+  # For some reason it's very important for the first cached nvcc invocation
+    cat > "/opt/cache/bin/$1" <<EOF
 #!/bin/sh
 
-# sccache does not support -E flag, so we need to call the original compiler directly in order to avoid calling this wrapper recursively
-for arg in "\$@"; do
-  if [ "\$arg" = "-E" ]; then
-    exec $(which $1) "\$@"
-  fi
-done
-
-if [ \$(env -u LD_PRELOAD ps -p \$PPID -o comm=) != sccache ]; then
+if [ "\$1" = "-E" ] || [ "\$2" = "-E" ]; then
+  exec $(which $1) "\$@"
+elif [ \$(env -u LD_PRELOAD ps -p \$PPID -o comm=) != sccache ]; then
   exec sccache $(which $1) "\$@"
 else
   exec $(which $1) "\$@"
 fi
 EOF
   else
-    cat >"/opt/cache/bin/$1" <<EOF
+    cat > "/opt/cache/bin/$1" <<EOF
 #!/bin/sh
 
 if [ \$(env -u LD_PRELOAD ps -p \$PPID -o comm=) != sccache ]; then
@@ -112,7 +125,7 @@ if [ -n "$ROCM_VERSION" ]; then
     TOPDIR=$(dirname $OLDCOMP)
     WRAPPED="$TOPDIR/original/$COMPNAME"
     mv "$OLDCOMP" "$WRAPPED"
-    printf "#!/bin/sh\nexec sccache $WRAPPED \"\$@\"" >"$OLDCOMP"
+    printf "#!/bin/sh\nexec sccache $WRAPPED \"\$@\"" > "$OLDCOMP"
     chmod a+x "$OLDCOMP"
   }
 

.ci/docker/common/install_conda.sh

@@ -25,8 +25,7 @@ if [ -n "$ANACONDA_PYTHON_VERSION" ]; then
   mkdir -p /opt/conda
   chown jenkins:jenkins /opt/conda
 
-  SCRIPT_FOLDER="$( cd "$(dirname "$0")" ; pwd -P )"
-  source "${SCRIPT_FOLDER}/common_utils.sh"
+  source "$(dirname "${BASH_SOURCE[0]}")/common_utils.sh"
 
   pushd /tmp
   wget -q "${BASE_URL}/${CONDA_FILE}"
@@ -85,9 +84,8 @@ if [ -n "$ANACONDA_PYTHON_VERSION" ]; then
 
   # Magma package names are concatenation of CUDA major and minor ignoring revision
   # I.e. magma-cuda102 package corresponds to CUDA_VERSION=10.2 and CUDA_VERSION=10.2.89
-  # Magma is installed from a tarball in the ossci-linux bucket into the conda env
   if [ -n "$CUDA_VERSION" ]; then
-    ${SCRIPT_FOLDER}/install_magma_conda.sh $(cut -f1-2 -d'.' <<< ${CUDA_VERSION}) ${ANACONDA_PYTHON_VERSION}
+    conda_install magma-cuda$(TMP=${CUDA_VERSION/./};echo ${TMP%.*[0-9]}) -c pytorch
   fi
 
   # Install some other packages, including those needed for Python test reporting

.ci/docker/common/install_cpython.sh

@@ -70,7 +70,7 @@ function do_cpython_build {
     # install setuptools since python 3.12 is required to use distutils
     ${prefix}/bin/pip install wheel==0.34.2 setuptools==68.2.2
     local abi_tag=$(${prefix}/bin/python -c "from wheel.pep425tags import get_abbr_impl, get_impl_ver, get_abi_tag; print('{0}{1}-{2}'.format(get_abbr_impl(), get_impl_ver(), get_abi_tag()))")
-    ln -sf ${prefix} /opt/python/${abi_tag}
+    ln -s ${prefix} /opt/python/${abi_tag}
 }
 
 function build_cpython {

.ci/docker/common/install_cuda.sh

@@ -2,7 +2,7 @@
 
 set -ex
 
-NCCL_VERSION=v2.25.1-1
+NCCL_VERSION=v2.21.5-1
 CUDNN_VERSION=9.5.1.17
 
 function install_cusparselt_040 {
@@ -16,6 +16,17 @@ function install_cusparselt_040 {
     rm -rf tmp_cusparselt
 }
 
+function install_cusparselt_052 {
+    # cuSparseLt license: https://docs.nvidia.com/cuda/cusparselt/license.html
+    mkdir tmp_cusparselt && pushd tmp_cusparselt
+    wget -q https://developer.download.nvidia.com/compute/cusparselt/redist/libcusparse_lt/linux-x86_64/libcusparse_lt-linux-x86_64-0.5.2.1-archive.tar.xz
+    tar xf libcusparse_lt-linux-x86_64-0.5.2.1-archive.tar.xz
+    cp -a libcusparse_lt-linux-x86_64-0.5.2.1-archive/include/* /usr/local/cuda/include/
+    cp -a libcusparse_lt-linux-x86_64-0.5.2.1-archive/lib/* /usr/local/cuda/lib64/
+    popd
+    rm -rf tmp_cusparselt
+}
+
 function install_cusparselt_062 {
     # cuSparseLt license: https://docs.nvidia.com/cuda/cusparselt/license.html
     mkdir tmp_cusparselt && pushd tmp_cusparselt
@@ -40,7 +51,6 @@ function install_cusparselt_063 {
 
 function install_118 {
     CUDNN_VERSION=9.1.0.70
-    NCCL_VERSION=v2.21.5-1
     echo "Installing CUDA 11.8 and cuDNN ${CUDNN_VERSION} and NCCL ${NCCL_VERSION} and cuSparseLt-0.4.0"
     rm -rf /usr/local/cuda-11.8 /usr/local/cuda
     # install CUDA 11.8.0 in the same container
@@ -73,6 +83,39 @@ function install_118 {
     ldconfig
 }
 
+function install_121 {
+    echo "Installing CUDA 12.1 and cuDNN ${CUDNN_VERSION} and NCCL ${NCCL_VERSION} and cuSparseLt-0.5.2"
+    rm -rf /usr/local/cuda-12.1 /usr/local/cuda
+    # install CUDA 12.1.0 in the same container
+    wget -q https://developer.download.nvidia.com/compute/cuda/12.1.1/local_installers/cuda_12.1.1_530.30.02_linux.run
+    chmod +x cuda_12.1.1_530.30.02_linux.run
+    ./cuda_12.1.1_530.30.02_linux.run --toolkit --silent
+    rm -f cuda_12.1.1_530.30.02_linux.run
+    rm -f /usr/local/cuda && ln -s /usr/local/cuda-12.1 /usr/local/cuda
+
+    # cuDNN license: https://developer.nvidia.com/cudnn/license_agreement
+    mkdir tmp_cudnn && cd tmp_cudnn
+    wget -q https://developer.download.nvidia.com/compute/cudnn/redist/cudnn/linux-x86_64/cudnn-linux-x86_64-${CUDNN_VERSION}_cuda12-archive.tar.xz -O cudnn-linux-x86_64-${CUDNN_VERSION}_cuda12-archive.tar.xz
+    tar xf cudnn-linux-x86_64-${CUDNN_VERSION}_cuda12-archive.tar.xz
+    cp -a cudnn-linux-x86_64-${CUDNN_VERSION}_cuda12-archive/include/* /usr/local/cuda/include/
+    cp -a cudnn-linux-x86_64-${CUDNN_VERSION}_cuda12-archive/lib/* /usr/local/cuda/lib64/
+    cd ..
+    rm -rf tmp_cudnn
+
+    # NCCL license: https://docs.nvidia.com/deeplearning/nccl/#licenses
+    # Follow build: https://github.com/NVIDIA/nccl/tree/master?tab=readme-ov-file#build
+    git clone -b $NCCL_VERSION --depth 1 https://github.com/NVIDIA/nccl.git
+    cd nccl && make -j src.build
+    cp -a build/include/* /usr/local/cuda/include/
+    cp -a build/lib/* /usr/local/cuda/lib64/
+    cd ..
+    rm -rf nccl
+
+    install_cusparselt_052
+
+    ldconfig
+}
+
 function install_124 {
   CUDNN_VERSION=9.1.0.70
   echo "Installing CUDA 12.4.1 and cuDNN ${CUDNN_VERSION} and NCCL ${NCCL_VERSION} and cuSparseLt-0.6.2"
@@ -171,6 +214,37 @@ function prune_118 {
     rm -rf $CUDA_BASE/libnvvp $CUDA_BASE/nsightee_plugins $CUDA_BASE/nsight-compute-2022.3.0 $CUDA_BASE/nsight-systems-2022.4.2/
 }
 
+function prune_121 {
+  echo "Pruning CUDA 12.1"
+  #####################################################################################
+  # CUDA 12.1 prune static libs
+  #####################################################################################
+    export NVPRUNE="/usr/local/cuda-12.1/bin/nvprune"
+    export CUDA_LIB_DIR="/usr/local/cuda-12.1/lib64"
+
+    export GENCODE="-gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_75,code=sm_75 -gencode arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -gencode arch=compute_90,code=sm_90"
+    export GENCODE_CUDNN="-gencode arch=compute_50,code=sm_50 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_75,code=sm_75 -gencode arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -gencode arch=compute_90,code=sm_90"
+
+    if [[ -n "$OVERRIDE_GENCODE" ]]; then
+        export GENCODE=$OVERRIDE_GENCODE
+    fi
+
+    # all CUDA libs except CuDNN and CuBLAS
+    ls $CUDA_LIB_DIR/ | grep "\.a" | grep -v "culibos" | grep -v "cudart" | grep -v "cudnn" | grep -v "cublas" | grep -v "metis"  \
+      | xargs -I {} bash -c \
+                "echo {} && $NVPRUNE $GENCODE $CUDA_LIB_DIR/{} -o $CUDA_LIB_DIR/{}"
+
+    # prune CuDNN and CuBLAS
+    $NVPRUNE $GENCODE_CUDNN $CUDA_LIB_DIR/libcublas_static.a -o $CUDA_LIB_DIR/libcublas_static.a
+    $NVPRUNE $GENCODE_CUDNN $CUDA_LIB_DIR/libcublasLt_static.a -o $CUDA_LIB_DIR/libcublasLt_static.a
+
+    #####################################################################################
+    # CUDA 12.1 prune visual tools
+    #####################################################################################
+    export CUDA_BASE="/usr/local/cuda-12.1/"
+    rm -rf $CUDA_BASE/libnvvp $CUDA_BASE/nsightee_plugins $CUDA_BASE/nsight-compute-2023.1.0 $CUDA_BASE/nsight-systems-2023.1.2/
+}
+
 function prune_124 {
   echo "Pruning CUDA 12.4"
   #####################################################################################
@@ -239,52 +313,18 @@ function prune_126 {
   rm -rf $CUDA_BASE/libnvvp $CUDA_BASE/nsightee_plugins $CUDA_BASE/nsight-compute-2024.3.2 $CUDA_BASE/nsight-systems-2024.5.1/
 }
 
-function install_128 {
-  CUDNN_VERSION=9.7.1.26
-  echo "Installing CUDA 12.8.0 and cuDNN ${CUDNN_VERSION} and NCCL ${NCCL_VERSION} and cuSparseLt-0.6.3"
-  rm -rf /usr/local/cuda-12.8 /usr/local/cuda
-  # install CUDA 12.8.0 in the same container
-  wget -q https://developer.download.nvidia.com/compute/cuda/12.8.0/local_installers/cuda_12.8.0_570.86.10_linux.run
-  chmod +x cuda_12.8.0_570.86.10_linux.run
-  ./cuda_12.8.0_570.86.10_linux.run --toolkit --silent
-  rm -f cuda_12.8.0_570.86.10_linux.run
-  rm -f /usr/local/cuda && ln -s /usr/local/cuda-12.8 /usr/local/cuda
-
-  # cuDNN license: https://developer.nvidia.com/cudnn/license_agreement
-  mkdir tmp_cudnn && cd tmp_cudnn
-  wget -q https://developer.download.nvidia.com/compute/cudnn/redist/cudnn/linux-x86_64/cudnn-linux-x86_64-${CUDNN_VERSION}_cuda12-archive.tar.xz -O cudnn-linux-x86_64-${CUDNN_VERSION}_cuda12-archive.tar.xz
-  tar xf cudnn-linux-x86_64-${CUDNN_VERSION}_cuda12-archive.tar.xz
-  cp -a cudnn-linux-x86_64-${CUDNN_VERSION}_cuda12-archive/include/* /usr/local/cuda/include/
-  cp -a cudnn-linux-x86_64-${CUDNN_VERSION}_cuda12-archive/lib/* /usr/local/cuda/lib64/
-  cd ..
-  rm -rf tmp_cudnn
-
-  # NCCL license: https://docs.nvidia.com/deeplearning/nccl/#licenses
-  # Follow build: https://github.com/NVIDIA/nccl/tree/master?tab=readme-ov-file#build
-  git clone -b $NCCL_VERSION --depth 1 https://github.com/NVIDIA/nccl.git
-  cd nccl && make -j src.build
-  cp -a build/include/* /usr/local/cuda/include/
-  cp -a build/lib/* /usr/local/cuda/lib64/
-  cd ..
-  rm -rf nccl
-
-  install_cusparselt_063
-
-  ldconfig
-}
-
 # idiomatic parameter and option handling in sh
 while test $# -gt 0
 do
     case "$1" in
     11.8) install_118; prune_118
         ;;
+    12.1) install_121; prune_121
+        ;;
     12.4) install_124; prune_124
         ;;
     12.6) install_126; prune_126
         ;;
-    12.8) install_128;
-        ;;
     *) echo "bad argument $1"; exit 1
         ;;
     esac

.ci/docker/common/install_cuda_aarch64.sh

@@ -20,10 +20,10 @@ function install_cusparselt_062 {
 function install_cusparselt_063 {
     # cuSparseLt license: https://docs.nvidia.com/cuda/cusparselt/license.html
     mkdir tmp_cusparselt && pushd tmp_cusparselt
-    wget -q https://developer.download.nvidia.com/compute/cusparselt/redist/libcusparse_lt/linux-sbsa/libcusparse_lt-linux-sbsa-0.6.3.2-archive.tar.xz
-    tar xf libcusparse_lt-linux-sbsa-0.6.3.2-archive.tar.xz
-    cp -a libcusparse_lt-linux-sbsa-0.6.3.2-archive/include/* /usr/local/cuda/include/
-    cp -a libcusparse_lt-linux-sbsa-0.6.3.2-archive/lib/* /usr/local/cuda/lib64/
+    wget -q https://developer.download.nvidia.com/compute/cusparselt/redist/libcusparse_lt/linux-x86_64/libcusparse_lt-linux-x86_64-0.6.3.2-archive.tar.xz
+    tar xf libcusparse_lt-linux-x86_64-0.6.3.2-archive.tar.xz
+    cp -a libcusparse_lt-linux-x86_64-0.6.3.2-archive/include/* /usr/local/cuda/include/
+    cp -a libcusparse_lt-linux-x86_64-0.6.3.2-archive/lib/* /usr/local/cuda/lib64/
     popd
     rm -rf tmp_cusparselt
 }
@@ -57,7 +57,7 @@ function install_124 {
   cd ..
   rm -rf nccl
 
-  install_cusparselt_063
+  install_cusparselt_062
 
   ldconfig
 }
@@ -160,40 +160,6 @@ function prune_126 {
   rm -rf $CUDA_BASE/libnvvp $CUDA_BASE/nsightee_plugins $CUDA_BASE/nsight-compute-2024.3.2 $CUDA_BASE/nsight-systems-2024.5.1/
 }
 
-function install_128 {
-  CUDNN_VERSION=9.7.1.26
-  echo "Installing CUDA 12.8.0 and cuDNN ${CUDNN_VERSION} and NCCL ${NCCL_VERSION} and cuSparseLt-0.6.3"
-  rm -rf /usr/local/cuda-12.8 /usr/local/cuda
-  # install CUDA 12.8.0 in the same container
-  wget -q https://developer.download.nvidia.com/compute/cuda/12.8.0/local_installers/cuda_12.8.0_570.86.10_linux_sbsa.run
-  chmod +x cuda_12.8.0_570.86.10_linux_sbsa.run
-  ./cuda_12.8.0_570.86.10_linux_sbsa.run --toolkit --silent
-  rm -f cuda_12.8.0_570.86.10_linux_sbsa.run
-  rm -f /usr/local/cuda && ln -s /usr/local/cuda-12.8 /usr/local/cuda
-
-  # cuDNN license: https://developer.nvidia.com/cudnn/license_agreement
-  mkdir tmp_cudnn && cd tmp_cudnn
-  wget -q https://developer.download.nvidia.com/compute/cudnn/redist/cudnn/linux-sbsa/cudnn-linux-sbsa-${CUDNN_VERSION}_cuda12-archive.tar.xz -O cudnn-linux-sbsa-${CUDNN_VERSION}_cuda12-archive.tar.xz
-  tar xf cudnn-linux-sbsa-${CUDNN_VERSION}_cuda12-archive.tar.xz
-  cp -a cudnn-linux-sbsa-${CUDNN_VERSION}_cuda12-archive/include/* /usr/local/cuda/include/
-  cp -a cudnn-linux-sbsa-${CUDNN_VERSION}_cuda12-archive/lib/* /usr/local/cuda/lib64/
-  cd ..
-  rm -rf tmp_cudnn
-
-  # NCCL license: https://docs.nvidia.com/deeplearning/nccl/#licenses
-  # Follow build: https://github.com/NVIDIA/nccl/tree/master?tab=readme-ov-file#build
-  git clone -b ${NCCL_VERSION} --depth 1 https://github.com/NVIDIA/nccl.git
-  cd nccl && make -j src.build
-  cp -a build/include/* /usr/local/cuda/include/
-  cp -a build/lib/* /usr/local/cuda/lib64/
-  cd ..
-  rm -rf nccl
-
-  install_cusparselt_063
-
-  ldconfig
-}
-
 # idiomatic parameter and option handling in sh
 while test $# -gt 0
 do
@@ -202,8 +168,6 @@ do
         ;;
     12.6) install_126; prune_126
         ;;
-    12.8) install_128;
-        ;;
     *) echo "bad argument $1"; exit 1
         ;;
     esac

.ci/docker/common/install_cudnn.sh

@@ -4,9 +4,7 @@ if [[ -n "${CUDNN_VERSION}" ]]; then
     # cuDNN license: https://developer.nvidia.com/cudnn/license_agreement
     mkdir tmp_cudnn
     pushd tmp_cudnn
-    if [[ ${CUDA_VERSION:0:4} == "12.8" ]]; then
-        CUDNN_NAME="cudnn-linux-x86_64-9.7.1.26_cuda12-archive"
-    elif [[ ${CUDA_VERSION:0:4} == "12.6" ]]; then
+    if [[ ${CUDA_VERSION:0:4} == "12.6" ]]; then
         CUDNN_NAME="cudnn-linux-x86_64-9.5.1.17_cuda12-archive"
     elif [[ ${CUDA_VERSION:0:2} == "12" ]]; then
         CUDNN_NAME="cudnn-linux-x86_64-9.1.0.70_cuda12-archive"

.ci/docker/common/install_cusparselt.sh

@@ -5,15 +5,7 @@ set -ex
 # cuSPARSELt license: https://docs.nvidia.com/cuda/cusparselt/license.html
 mkdir tmp_cusparselt && cd tmp_cusparselt
 
-if [[ ${CUDA_VERSION:0:4} =~ ^12\.[5-8]$ ]]; then
-    arch_path='sbsa'
-    export TARGETARCH=${TARGETARCH:-$(uname -m)}
-    if [ ${TARGETARCH} = 'amd64' ] || [ "${TARGETARCH}" = 'x86_64' ]; then
-        arch_path='x86_64'
-    fi
-    CUSPARSELT_NAME="libcusparse_lt-linux-${arch_path}-0.6.3.2-archive"
-    curl --retry 3 -OLs https://developer.download.nvidia.com/compute/cusparselt/redist/libcusparse_lt/linux-${arch_path}/${CUSPARSELT_NAME}.tar.xz
-elif [[ ${CUDA_VERSION:0:4} == "12.4" ]]; then
+if [[ ${CUDA_VERSION:0:4} =~ ^12\.[2-6]$ ]]; then
     arch_path='sbsa'
     export TARGETARCH=${TARGETARCH:-$(uname -m)}
     if [ ${TARGETARCH} = 'amd64' ] || [ "${TARGETARCH}" = 'x86_64' ]; then
@@ -21,11 +13,17 @@ elif [[ ${CUDA_VERSION:0:4} == "12.4" ]]; then
     fi
     CUSPARSELT_NAME="libcusparse_lt-linux-${arch_path}-0.6.2.3-archive"
     curl --retry 3 -OLs https://developer.download.nvidia.com/compute/cusparselt/redist/libcusparse_lt/linux-${arch_path}/${CUSPARSELT_NAME}.tar.xz
+elif [[ ${CUDA_VERSION:0:4} == "12.1" ]]; then
+    arch_path='sbsa'
+    export TARGETARCH=${TARGETARCH:-$(uname -m)}
+    if [ ${TARGETARCH} = 'amd64' ] || [ "${TARGETARCH}" = 'x86_64' ]; then
+        arch_path='x86_64'
+    fi
+    CUSPARSELT_NAME="libcusparse_lt-linux-${arch_path}-0.5.2.1-archive"
+    curl --retry 3 -OLs https://developer.download.nvidia.com/compute/cusparselt/redist/libcusparse_lt/linux-${arch_path}/${CUSPARSELT_NAME}.tar.xz
 elif [[ ${CUDA_VERSION:0:4} == "11.8" ]]; then
     CUSPARSELT_NAME="libcusparse_lt-linux-x86_64-0.4.0.7-archive"
     curl --retry 3 -OLs https://developer.download.nvidia.com/compute/cusparselt/redist/libcusparse_lt/linux-x86_64/${CUSPARSELT_NAME}.tar.xz
-else
-    echo "Not sure which libcusparselt version to install for this ${CUDA_VERSION}"
 fi
 
 tar xf ${CUSPARSELT_NAME}.tar.xz

.ci/docker/common/install_executorch.sh

@@ -37,17 +37,14 @@ install_conda_dependencies() {
 
 install_pip_dependencies() {
   pushd executorch
-  as_jenkins bash install_executorch.sh
-
-  # A workaround, ExecuTorch has moved to numpy 2.0 which is not compatible with the current
-  # numba and scipy version used in PyTorch CI
-  conda_run pip uninstall -y numba scipy
-
+  as_jenkins bash install_requirements.sh --pybind xnnpack
   popd
 }
 
 setup_executorch() {
   pushd executorch
+  # Setup swiftshader and Vulkan SDK which are required to build the Vulkan delegate
+  as_jenkins bash .ci/scripts/setup-vulkan-linux-deps.sh
 
   export PYTHON_EXECUTABLE=python
   export EXECUTORCH_BUILD_PYBIND=ON

.ci/docker/common/install_inductor_benchmark_deps.sh

@@ -7,20 +7,14 @@ source "$(dirname "${BASH_SOURCE[0]}")/common_utils.sh"
 function install_huggingface() {
   local version
   commit=$(get_pinned_commit huggingface)
+  pip_install pandas==2.0.3
   pip_install "git+https://github.com/huggingface/transformers@${commit}"
 }
 
 function install_timm() {
   local commit
   commit=$(get_pinned_commit timm)
-
-  # TODO (huydhn): There is no torchvision release on 3.13 when I write this, so
-  # I'm using nightly here instead. We just need to package to be able to install
-  # TIMM. Removing this once vision has a release on 3.13
-  if [[ "${ANACONDA_PYTHON_VERSION}" == "3.13" ]]; then
-    pip_install --pre torch torchvision --index-url https://download.pytorch.org/whl/nightly/cu124
-  fi
-
+  pip_install pandas==2.0.3
   pip_install "git+https://github.com/huggingface/pytorch-image-models@${commit}"
   # Clean up
   conda_run pip uninstall -y cmake torch torchvision triton

.ci/docker/common/install_magma.sh

@@ -3,6 +3,8 @@
 
 set -eou pipefail
 
+MAGMA_VERSION="2.5.2"
+
 function do_install() {
     cuda_version=$1
     cuda_version_nodot=${1/./}
@@ -15,7 +17,7 @@ function do_install() {
         set -x
         tmp_dir=$(mktemp -d)
         pushd ${tmp_dir}
-        curl -OLs https://ossci-linux.s3.us-east-1.amazonaws.com/${magma_archive}
+        curl -OLs https://anaconda.org/pytorch/magma-cuda${cuda_version_nodot}/${MAGMA_VERSION}/download/linux-64/${magma_archive}
         tar -xvf "${magma_archive}"
         mkdir -p "${cuda_dir}/magma"
         mv include "${cuda_dir}/magma/include"

.ci/docker/common/install_magma_conda.sh

@@ -1,26 +0,0 @@
-#!/usr/bin/env bash
-# Script that replaces the magma install from a conda package
-
-set -eou pipefail
-
-function do_install() {
-    cuda_version_nodot=${1/./}
-    anaconda_python_version=$2
-
-    MAGMA_VERSION="2.6.1"
-    magma_archive="magma-cuda${cuda_version_nodot}-${MAGMA_VERSION}-1.tar.bz2"
-
-    anaconda_dir="/opt/conda/envs/py_${anaconda_python_version}"
-    (
-        set -x
-        tmp_dir=$(mktemp -d)
-        pushd ${tmp_dir}
-        curl -OLs https://ossci-linux.s3.us-east-1.amazonaws.com/${magma_archive}
-        tar -xvf "${magma_archive}"
-        mv include/* "${anaconda_dir}/include/"
-        mv lib/* "${anaconda_dir}/lib"
-        popd
-    )
-}
-
-do_install $1 $2

.ci/docker/common/install_onnx.sh

@@ -31,15 +31,15 @@ pip_install \
 pip_install coloredlogs packaging
 
 pip_install onnxruntime==1.18.1
-pip_install onnx==1.17.0
-pip_install onnxscript==0.1.0 --no-deps
+pip_install onnx==1.16.2
+pip_install onnxscript==0.1.0.dev20241124 --no-deps
 # required by onnxscript
 pip_install ml_dtypes
 
 # Cache the transformers model to be used later by ONNX tests. We need to run the transformers
 # package to download the model. By default, the model is cached at ~/.cache/huggingface/hub/
 IMPORT_SCRIPT_FILENAME="/tmp/onnx_import_script.py"
-as_jenkins echo 'import transformers; transformers.GPTJForCausalLM.from_pretrained("hf-internal-testing/tiny-random-gptj");' > "${IMPORT_SCRIPT_FILENAME}"
+as_jenkins echo 'import transformers; transformers.AutoModel.from_pretrained("sshleifer/tiny-gpt2"); transformers.AutoTokenizer.from_pretrained("sshleifer/tiny-gpt2"); transformers.AutoModelForSpeechSeq2Seq.from_pretrained("openai/whisper-large-v3");' > "${IMPORT_SCRIPT_FILENAME}"
 
 # Need a PyTorch version for transformers to work
 pip_install --pre torch --index-url https://download.pytorch.org/whl/nightly/cpu

.ci/docker/common/install_rocm.sh

@@ -62,22 +62,6 @@ install_ubuntu() {
         sqlite3 $kdb "PRAGMA journal_mode=off; PRAGMA VACUUM;"
     done
 
-    # ROCm 6.3 had a regression where initializing static code objects had significant overhead
-    if [[ $(ver $ROCM_VERSION) -eq $(ver 6.3) ]]; then
-        # clr build needs CppHeaderParser but can only find it using conda's python
-        /opt/conda/bin/python -m pip install CppHeaderParser
-        git clone https://github.com/ROCm/HIP -b rocm-6.3.x
-        HIP_COMMON_DIR=$(readlink -f HIP)
-        git clone https://github.com/jeffdaily/clr -b release/rocm-rel-6.3-statco-hotfix
-        mkdir -p clr/build
-        pushd clr/build
-        cmake .. -DCLR_BUILD_HIP=ON -DHIP_COMMON_DIR=$HIP_COMMON_DIR
-        make -j
-        cp hipamd/lib/libamdhip64.so.6.3.* /opt/rocm/lib/libamdhip64.so.6.3.*
-        popd
-        rm -rf HIP clr
-    fi
-
     # Cleanup
     apt-get autoclean && apt-get clean
     rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*

.ci/docker/common/install_ucc.sh

@@ -8,12 +8,6 @@ else
   with_cuda=no
 fi
 
-if [[ -d "/opt/rocm" ]]; then
-  with_rocm=/opt/rocm
-else
-  with_rocm=no
-fi
-
 function install_ucx() {
   set -ex
   git clone --recursive https://github.com/openucx/ucx.git
@@ -25,7 +19,6 @@ function install_ucx() {
   ./configure --prefix=$UCX_HOME      \
       --enable-mt                     \
       --with-cuda=$with_cuda          \
-      --with-rocm=$with_rocm          \
       --enable-profiling              \
       --enable-stats
   time make -j
@@ -43,29 +36,12 @@ function install_ucc() {
   git submodule update --init --recursive
 
   ./autogen.sh
-
   # We only run distributed tests on Tesla M60 and A10G
   NVCC_GENCODE="-gencode=arch=compute_52,code=sm_52 -gencode=arch=compute_86,code=compute_86"
-
-  if [[ -n "$ROCM_VERSION" ]]; then
-    if [[ -n "$PYTORCH_ROCM_ARCH" ]]; then
-      amdgpu_targets=`echo $PYTORCH_ROCM_ARCH | sed 's/;/ /g'`
-    else
-      amdgpu_targets=`rocm_agent_enumerator | grep -v gfx000 | sort -u | xargs`
-    fi
-    for arch in $amdgpu_targets; do
-      HIP_OFFLOAD="$HIP_OFFLOAD --offload-arch=$arch"
-    done
-  else
-    HIP_OFFLOAD="all-arch-no-native"
-  fi
-
   ./configure --prefix=$UCC_HOME          \
     --with-ucx=$UCX_HOME                  \
     --with-cuda=$with_cuda                \
-    --with-nvcc-gencode="${NVCC_GENCODE}" \
-    --with-rocm=$with_rocm                \
-    --with-rocm-arch="${HIP_OFFLOAD}"
+    --with-nvcc-gencode="${NVCC_GENCODE}"
   time make -j
   sudo make install
 

.ci/docker/libtorch/Dockerfile

@@ -56,6 +56,11 @@ RUN bash ./install_cuda.sh 11.8
 RUN bash ./install_magma.sh 11.8
 RUN ln -sf /usr/local/cuda-11.8 /usr/local/cuda
 
+FROM cuda as cuda12.1
+RUN bash ./install_cuda.sh 12.1
+RUN bash ./install_magma.sh 12.1
+RUN ln -sf /usr/local/cuda-12.1 /usr/local/cuda
+
 FROM cuda as cuda12.4
 RUN bash ./install_cuda.sh 12.4
 RUN bash ./install_magma.sh 12.4
@@ -66,11 +71,6 @@ RUN bash ./install_cuda.sh 12.6
 RUN bash ./install_magma.sh 12.6
 RUN ln -sf /usr/local/cuda-12.6 /usr/local/cuda
 
-FROM cuda as cuda12.8
-RUN bash ./install_cuda.sh 12.8
-RUN bash ./install_magma.sh 12.8
-RUN ln -sf /usr/local/cuda-12.8 /usr/local/cuda
-
 FROM cpu as rocm
 ARG PYTORCH_ROCM_ARCH
 ENV PYTORCH_ROCM_ARCH ${PYTORCH_ROCM_ARCH}
@@ -92,6 +92,13 @@ RUN apt-get update -y && \
 RUN bash ./install_rocm_drm.sh && rm install_rocm_drm.sh
 RUN bash ./install_rocm_magma.sh && rm install_rocm_magma.sh
 
+# Install AOTriton
+COPY ./common/common_utils.sh common_utils.sh
+COPY ./aotriton_version.txt aotriton_version.txt
+COPY ./common/install_aotriton.sh install_aotriton.sh
+RUN bash ./install_aotriton.sh /opt/rocm && rm install_aotriton.sh aotriton_version.txt
+ENV AOTRITON_INSTALLED_PREFIX /opt/rocm/aotriton
+
 FROM ${BASE_TARGET} as final
 COPY --from=openssl            /opt/openssl           /opt/openssl
 # Install patchelf

.ci/docker/libtorch/build.sh

@@ -39,7 +39,17 @@ case ${GPU_ARCH_TYPE} in
         BASE_TARGET=rocm
         DOCKER_TAG=rocm${GPU_ARCH_VERSION}
         GPU_IMAGE=rocm/dev-ubuntu-20.04:${GPU_ARCH_VERSION}-complete
-        PYTORCH_ROCM_ARCH="gfx900;gfx906;gfx908;gfx90a;gfx1030;gfx1100;gfx1101;gfx942"
+        PYTORCH_ROCM_ARCH="gfx900;gfx906;gfx908;gfx90a;gfx1030;gfx1100"
+        ROCM_REGEX="([0-9]+)\.([0-9]+)[\.]?([0-9]*)"
+        if [[ $GPU_ARCH_VERSION =~ $ROCM_REGEX ]]; then
+            ROCM_VERSION_INT=$((${BASH_REMATCH[1]}*10000 + ${BASH_REMATCH[2]}*100 + ${BASH_REMATCH[3]:-0}))
+        else
+            echo "ERROR: rocm regex failed"
+            exit 1
+        fi
+        if [[ $ROCM_VERSION_INT -ge 60000 ]]; then
+            PYTORCH_ROCM_ARCH+=";gfx942"
+        fi
         DOCKER_GPU_BUILD_ARG="--build-arg PYTORCH_ROCM_ARCH=${PYTORCH_ROCM_ARCH}"
         ;;
     *)

.ci/docker/linter-cuda/Dockerfile

@@ -25,8 +25,7 @@ ENV PATH /opt/conda/envs/py_$ANACONDA_PYTHON_VERSION/bin:/opt/conda/bin:$PATH
 COPY requirements-ci.txt /opt/conda/requirements-ci.txt
 COPY ./common/install_conda.sh install_conda.sh
 COPY ./common/common_utils.sh common_utils.sh
-COPY ./common/install_magma_conda.sh install_magma_conda.sh
-RUN bash ./install_conda.sh && rm install_conda.sh install_magma_conda.sh common_utils.sh /opt/conda/requirements-ci.txt
+RUN bash ./install_conda.sh && rm install_conda.sh common_utils.sh /opt/conda/requirements-ci.txt
 
 # Install cuda and cudnn
 ARG CUDA_VERSION

.ci/docker/manywheel/Dockerfile

@@ -198,3 +198,10 @@ ADD ./common/install_rocm_magma.sh install_rocm_magma.sh
 RUN bash ./install_rocm_magma.sh && rm install_rocm_magma.sh
 ADD ./common/install_miopen.sh install_miopen.sh
 RUN bash ./install_miopen.sh ${ROCM_VERSION} && rm install_miopen.sh
+
+# Install AOTriton
+COPY ./common/common_utils.sh common_utils.sh
+COPY ./aotriton_version.txt aotriton_version.txt
+COPY ./common/install_aotriton.sh install_aotriton.sh
+RUN bash ./install_aotriton.sh /opt/rocm && rm install_aotriton.sh aotriton_version.txt
+ENV AOTRITON_INSTALLED_PREFIX /opt/rocm/aotriton

.ci/docker/manywheel/build.sh

@@ -97,7 +97,14 @@ case ${GPU_ARCH_TYPE} in
             DEVTOOLSET_VERSION="11"
             GPU_IMAGE=rocm/dev-almalinux-8:${GPU_ARCH_VERSION}-complete
         fi
-        PYTORCH_ROCM_ARCH="gfx900;gfx906;gfx908;gfx90a;gfx942;gfx1030;gfx1100;gfx1101"
+        PYTORCH_ROCM_ARCH="gfx900;gfx906;gfx908;gfx90a;gfx942;gfx1030;gfx1100"
+        ROCM_REGEX="([0-9]+)\.([0-9]+)[\.]?([0-9]*)"
+        if [[ $GPU_ARCH_VERSION =~ $ROCM_REGEX ]]; then
+            ROCM_VERSION_INT=$((${BASH_REMATCH[1]}*10000 + ${BASH_REMATCH[2]}*100 + ${BASH_REMATCH[3]:-0}))
+        else
+            echo "ERROR: rocm regex failed"
+            exit 1
+        fi
         DOCKER_GPU_BUILD_ARG="--build-arg ROCM_VERSION=${GPU_ARCH_VERSION} --build-arg PYTORCH_ROCM_ARCH=${PYTORCH_ROCM_ARCH} --build-arg DEVTOOLSET_VERSION=${DEVTOOLSET_VERSION}"
         ;;
     xpu)

.ci/docker/requirements-ci.txt

@@ -30,10 +30,10 @@ dill==0.3.7
 #Pinned versions: 0.3.7
 #test that import: dynamo/test_replay_record.py test_dataloader.py test_datapipe.py test_serialization.py
 
-expecttest==0.3.0
+expecttest==0.2.1
 #Description: method for writing tests where test framework auto populates
 # the expected output based on previous runs
-#Pinned versions: 0.3.0
+#Pinned versions: 0.2.1
 #test that import:
 
 fbscribelogger==0.1.7
@@ -90,7 +90,7 @@ librosa>=0.6.2 ; python_version < "3.11"
 #Pinned versions:
 #test that import:
 
-mypy==1.13.0
+mypy==1.11.2
 # Pin MyPy version because new errors are likely to appear with each release
 #Description: linter
 #Pinned versions: 1.10.0
@@ -132,9 +132,6 @@ numpy==1.22.4; python_version == "3.9" or python_version == "3.10"
 numpy==1.26.2; python_version == "3.11" or python_version == "3.12"
 numpy==2.1.2; python_version >= "3.13"
 
-pandas==2.0.3; python_version < "3.13"
-pandas==2.2.3; python_version >= "3.13"
-
 #onnxruntime
 #Description: scoring engine for Open Neural Network Exchange (ONNX) models
 #Pinned versions: 1.9.0
@@ -158,7 +155,7 @@ optree==0.13.0
 #test_pointwise_ops.py, test_dtensor_ops.py, test_torchinductor.py, test_fx.py,
 #test_fake_tensor.py, test_mps.py
 
-pillow==11.0.0
+pillow==10.3.0
 #Description:  Python Imaging Library fork
 #Pinned versions: 10.3.0
 #test that import:
@@ -193,11 +190,6 @@ pytest-rerunfailures>=10.3
 #Pinned versions:
 #test that import:
 
-pytest-subtests==0.13.1
-#Description: plugin for subtest support
-#Pinned versions:
-#test that import:
-
 #pytest-benchmark
 #Description: fixture for benchmarking code
 #Pinned versions: 3.2.3
@@ -245,7 +237,7 @@ scikit-image==0.22.0 ; python_version >= "3.10"
 #test that import:
 
 scipy==1.10.1 ; python_version <= "3.11"
-scipy==1.14.1 ; python_version >= "3.12"
+scipy==1.12.0 ; python_version == "3.12"
 # Pin SciPy because of failing distribution tests (see #60347)
 #Description: scientific python
 #Pinned versions: 1.10.1
@@ -280,9 +272,9 @@ unittest-xml-reporting<=3.2.0,>=2.0.0
 #test that import:
 
 #lintrunner is supported on aarch64-linux only from 0.12.4 version
-lintrunner==0.12.7
+lintrunner==0.12.5
 #Description: all about linters!
-#Pinned versions: 0.12.7
+#Pinned versions: 0.12.5
 #test that import:
 
 redis>=4.0.0
@@ -294,7 +286,7 @@ ghstack==0.8.0
 #Pinned versions: 0.8.0
 #test that import:
 
-jinja2==3.1.5
+jinja2==3.1.4
 #Description: jinja2 template engine
 #Pinned versions: 3.1.4
 #test that import:
@@ -304,32 +296,31 @@ pytest-cpp==2.3.0
 #Pinned versions: 2.3.0
 #test that import:
 
-z3-solver==4.12.6.0
+z3-solver==4.12.2.0
 #Description: The Z3 Theorem Prover Project
 #Pinned versions:
 #test that import:
 
-tensorboard==2.13.0 ; python_version < "3.13"
-tensorboard==2.18.0 ; python_version >= "3.13"
+tensorboard==2.13.0
 #Description: Also included in .ci/docker/requirements-docs.txt
 #Pinned versions:
 #test that import: test_tensorboard
 
 pywavelets==1.4.1 ; python_version < "3.12"
-pywavelets==1.7.0 ; python_version >= "3.12"
+pywavelets==1.5.0 ; python_version >= "3.12"
 #Description: This is a requirement of scikit-image, we need to pin
 # it here because 1.5.0 conflicts with numpy 1.21.2 used in CI
 #Pinned versions: 1.4.1
 #test that import:
 
-lxml==5.3.0
+lxml==5.0.0
 #Description: This is a requirement of unittest-xml-reporting
 
 # Python-3.9 binaries
 
 PyGithub==2.3.0
 
-sympy==1.13.3
+sympy==1.13.1 ; python_version >= "3.9"
 #Description: Required by coremltools, also pinned in .github/requirements/pip-requirements-macOS.txt
 #Pinned versions:
 #test that import:
@@ -339,7 +330,7 @@ onnx==1.17.0
 #Pinned versions:
 #test that import:
 
-onnxscript==0.1.0
+onnxscript==0.1.0.dev20240817
 #Description: Required by mypy and test_public_bindings.py when checking torch.onnx._internal
 #Pinned versions:
 #test that import:
@@ -362,7 +353,6 @@ pwlf==2.2.1 ; python_version >= "3.8"
 # To build PyTorch itself
 astunparse
 PyYAML
-pyzstd
 setuptools
 
 ninja==1.11.1 ; platform_machine == "aarch64"
@@ -372,8 +362,3 @@ pulp==2.9.0 ; python_version >= "3.8"
 #Description: required for testing ilp formulaiton under torch/distributed/_tools
 #Pinned versions: 2.9.0
 #test that import: test_sac_ilp.py
-
-dataclasses_json==0.6.7
-#Description: required for data pipeline and scripts under tools/stats
-#Pinned versions: 0.6.7
-#test that import:

.ci/docker/requirements-docs.txt

@@ -14,8 +14,7 @@ matplotlib==3.5.3
 #Description: This is used to generate PyTorch docs
 #Pinned versions: 3.5.3
 
-tensorboard==2.13.0 ; python_version < "3.13"
-tensorboard==2.18.0 ; python_version >= "3.13"
+tensorboard==2.13.0
 #Description: This is used to generate PyTorch docs
 #Pinned versions: 2.13.0
 

.ci/docker/ubuntu-cuda/Dockerfile

@@ -30,8 +30,7 @@ ARG CONDA_CMAKE
 COPY requirements-ci.txt /opt/conda/requirements-ci.txt
 COPY ./common/install_conda.sh install_conda.sh
 COPY ./common/common_utils.sh common_utils.sh
-COPY ./common/install_magma_conda.sh install_magma_conda.sh
-RUN bash ./install_conda.sh && rm install_conda.sh install_magma_conda.sh common_utils.sh /opt/conda/requirements-ci.txt
+RUN bash ./install_conda.sh && rm install_conda.sh common_utils.sh /opt/conda/requirements-ci.txt
 
 # Install gcc
 ARG GCC_VERSION
@@ -81,8 +80,6 @@ RUN bash ./install_openssl.sh
 ENV OPENSSL_DIR /opt/openssl
 
 ARG INDUCTOR_BENCHMARKS
-ARG ANACONDA_PYTHON_VERSION
-ENV ANACONDA_PYTHON_VERSION=$ANACONDA_PYTHON_VERSION
 COPY ./common/install_inductor_benchmark_deps.sh install_inductor_benchmark_deps.sh
 COPY ./common/common_utils.sh common_utils.sh
 COPY ci_commit_pins/huggingface.txt huggingface.txt

.ci/docker/ubuntu-rocm/Dockerfile

@@ -14,20 +14,21 @@ ENV PYTORCH_ROCM_ARCH ${PYTORCH_ROCM_ARCH}
 COPY ./common/install_base.sh install_base.sh
 RUN bash ./install_base.sh && rm install_base.sh
 
+# Install clang
+ARG LLVMDEV
+ARG CLANG_VERSION
+COPY ./common/install_clang.sh install_clang.sh
+RUN bash ./install_clang.sh && rm install_clang.sh
+
 # Install user
 COPY ./common/install_user.sh install_user.sh
 RUN bash ./install_user.sh && rm install_user.sh
 
-# Install katex
-ARG KATEX
-COPY ./common/install_docs_reqs.sh install_docs_reqs.sh
-RUN bash ./install_docs_reqs.sh && rm install_docs_reqs.sh
-
 # Install conda and other packages (e.g., numpy, pytest)
 ARG ANACONDA_PYTHON_VERSION
+ARG CONDA_CMAKE
 ENV ANACONDA_PYTHON_VERSION=$ANACONDA_PYTHON_VERSION
 ENV PATH /opt/conda/envs/py_$ANACONDA_PYTHON_VERSION/bin:/opt/conda/bin:$PATH
-ARG CONDA_CMAKE
 COPY requirements-ci.txt /opt/conda/requirements-ci.txt
 COPY ./common/install_conda.sh install_conda.sh
 COPY ./common/common_utils.sh common_utils.sh
@@ -38,11 +39,6 @@ ARG GCC_VERSION
 COPY ./common/install_gcc.sh install_gcc.sh
 RUN bash ./install_gcc.sh && rm install_gcc.sh
 
-# Install clang
-ARG CLANG_VERSION
-COPY ./common/install_clang.sh install_clang.sh
-RUN bash ./install_clang.sh && rm install_clang.sh
-
 # (optional) Install protobuf for ONNX
 ARG PROTOBUF
 COPY ./common/install_protobuf.sh install_protobuf.sh
@@ -89,32 +85,6 @@ COPY ./common/install_amdsmi.sh install_amdsmi.sh
 RUN bash ./install_amdsmi.sh
 RUN rm install_amdsmi.sh
 
-# (optional) Install UCC
-ARG UCX_COMMIT
-ARG UCC_COMMIT
-ENV UCX_COMMIT $UCX_COMMIT
-ENV UCC_COMMIT $UCC_COMMIT
-ENV UCX_HOME /usr
-ENV UCC_HOME /usr
-ADD ./common/install_ucc.sh install_ucc.sh
-RUN if [ -n "${UCX_COMMIT}" ] && [ -n "${UCC_COMMIT}" ]; then bash ./install_ucc.sh; fi
-RUN rm install_ucc.sh
-
-COPY ./common/install_openssl.sh install_openssl.sh
-ENV OPENSSL_ROOT_DIR /opt/openssl
-RUN bash ./install_openssl.sh
-ENV OPENSSL_DIR /opt/openssl
-
-ARG INDUCTOR_BENCHMARKS
-ARG ANACONDA_PYTHON_VERSION
-ENV ANACONDA_PYTHON_VERSION=$ANACONDA_PYTHON_VERSION
-COPY ./common/install_inductor_benchmark_deps.sh install_inductor_benchmark_deps.sh
-COPY ./common/common_utils.sh common_utils.sh
-COPY ci_commit_pins/huggingface.txt huggingface.txt
-COPY ci_commit_pins/timm.txt timm.txt
-RUN if [ -n "${INDUCTOR_BENCHMARKS}" ]; then bash ./install_inductor_benchmark_deps.sh; fi
-RUN rm install_inductor_benchmark_deps.sh common_utils.sh timm.txt huggingface.txt
-
 # (optional) Install non-default CMake version
 ARG CMAKE_VERSION
 COPY ./common/install_cmake.sh install_cmake.sh
@@ -137,17 +107,18 @@ COPY triton_version.txt triton_version.txt
 RUN if [ -n "${TRITON}" ]; then bash ./install_triton.sh; fi
 RUN rm install_triton.sh common_utils.sh triton.txt triton_version.txt
 
+# Install AOTriton
+COPY ./aotriton_version.txt aotriton_version.txt
+COPY ./common/common_utils.sh common_utils.sh
+COPY ./common/install_aotriton.sh install_aotriton.sh
+RUN ["/bin/bash", "-c", "./install_aotriton.sh /opt/rocm && rm -rf install_aotriton.sh aotriton_version.txt common_utils.sh"]
+ENV AOTRITON_INSTALLED_PREFIX /opt/rocm/aotriton
 
 # Install ccache/sccache (do this last, so we get priority in PATH)
 COPY ./common/install_cache.sh install_cache.sh
 ENV PATH /opt/cache/bin:$PATH
 RUN bash ./install_cache.sh && rm install_cache.sh
 
-# Install Open MPI for ROCm
-COPY ./common/install_openmpi.sh install_openmpi.sh
-RUN if [ -n "${CUDA_VERSION}" ]; then bash install_openmpi.sh; fi
-RUN rm install_openmpi.sh
-
 # Include BUILD_ENVIRONMENT environment variable in image
 ARG BUILD_ENVIRONMENT
 ENV BUILD_ENVIRONMENT ${BUILD_ENVIRONMENT}

.ci/docker/ubuntu/Dockerfile

@@ -36,8 +36,7 @@ ENV DOCS=$DOCS
 COPY requirements-ci.txt requirements-docs.txt /opt/conda/
 COPY ./common/install_conda.sh install_conda.sh
 COPY ./common/common_utils.sh common_utils.sh
-COPY ./common/install_magma_conda.sh install_magma_conda.sh
-RUN bash ./install_conda.sh && rm install_conda.sh install_magma_conda.sh common_utils.sh /opt/conda/requirements-ci.txt /opt/conda/requirements-docs.txt
+RUN bash ./install_conda.sh && rm install_conda.sh common_utils.sh /opt/conda/requirements-ci.txt /opt/conda/requirements-docs.txt
 RUN if [ -n "${UNINSTALL_DILL}" ]; then pip uninstall -y dill; fi
 
 # Install gcc

.ci/magma/Makefile

@@ -16,9 +16,9 @@ DOCKER_RUN = set -eou pipefail; ${DOCKER_CMD} run --rm -i \
 	magma/build_magma.sh
 
 .PHONY: all
-all: magma-cuda128
 all: magma-cuda126
 all: magma-cuda124
+all: magma-cuda121
 all: magma-cuda118
 
 .PHONY:
@@ -26,12 +26,6 @@ clean:
 	$(RM) -r magma-*
 	$(RM) -r output
 
-.PHONY: magma-cuda128
-magma-cuda128: DESIRED_CUDA := 12.8
-magma-cuda128: CUDA_ARCH_LIST += -gencode arch=compute_100,code=sm_100 -gencode arch=compute_120,code=sm_120
-magma-cuda128:
-	$(DOCKER_RUN)
-
 .PHONY: magma-cuda126
 magma-cuda126: DESIRED_CUDA := 12.6
 magma-cuda126:
@@ -42,6 +36,11 @@ magma-cuda124: DESIRED_CUDA := 12.4
 magma-cuda124:
 	$(DOCKER_RUN)
 
+.PHONY: magma-cuda121
+magma-cuda121: DESIRED_CUDA := 12.1
+magma-cuda121:
+	$(DOCKER_RUN)
+
 .PHONY: magma-cuda118
 magma-cuda118: DESIRED_CUDA := 11.8
 magma-cuda118: CUDA_ARCH_LIST += -gencode arch=compute_37,code=sm_37

.ci/manywheel/build_common.sh

@@ -18,14 +18,12 @@ retry () {
     $*  || (sleep 1 && $*) || (sleep 2 && $*) || (sleep 4 && $*) || (sleep 8 && $*)
 }
 
-PLATFORM="manylinux2014_x86_64"
 # TODO move this into the Docker images
 OS_NAME=$(awk -F= '/^NAME/{print $2}' /etc/os-release)
 if [[ "$OS_NAME" == *"CentOS Linux"* ]]; then
     retry yum install -q -y zip openssl
 elif [[ "$OS_NAME" == *"AlmaLinux"* ]]; then
     retry yum install -q -y zip openssl
-    PLATFORM="manylinux_2_28_x86_64"
 elif [[ "$OS_NAME" == *"Red Hat Enterprise Linux"* ]]; then
     retry dnf install -q -y zip openssl
 elif [[ "$OS_NAME" == *"Ubuntu"* ]]; then
@@ -255,11 +253,11 @@ make_wheel_record() {
     FPATH=$1
     if echo $FPATH | grep RECORD >/dev/null 2>&1; then
         # if the RECORD file, then
-        echo "\"$FPATH\",,"
+        echo "$FPATH,,"
     else
         HASH=$(openssl dgst -sha256 -binary $FPATH | openssl base64 | sed -e 's/+/-/g' | sed -e 's/\//_/g' | sed -e 's/=//g')
         FSIZE=$(ls -nl $FPATH | awk '{print $5}')
-        echo "\"$FPATH\",sha256=$HASH,$FSIZE"
+        echo "$FPATH,sha256=$HASH,$FSIZE"
     fi
 }
 
@@ -379,12 +377,6 @@ for pkg in /$WHEELHOUSE_DIR/torch_no_python*.whl /$WHEELHOUSE_DIR/torch*linux*.w
         $PATCHELF_BIN --print-rpath $sofile
     done
 
-    # create Manylinux 2_28 tag this needs to happen before regenerate the RECORD
-    if [[ $PLATFORM == "manylinux_2_28_x86_64" && $GPU_ARCH_TYPE != "cpu-s390x" && $GPU_ARCH_TYPE != "xpu" ]]; then
-        wheel_file=$(echo $(basename $pkg) | sed -e 's/-cp.*$/.dist-info\/WHEEL/g')
-        sed -i -e s#linux_x86_64#"${PLATFORM}"# $wheel_file;
-    fi
-
     # regenerate the RECORD file with new hashes
     record_file=$(echo $(basename $pkg) | sed -e 's/-cp.*$/.dist-info\/RECORD/g')
     if [[ -e $record_file ]]; then
@@ -424,20 +416,12 @@ for pkg in /$WHEELHOUSE_DIR/torch_no_python*.whl /$WHEELHOUSE_DIR/torch*linux*.w
         popd
     fi
 
-    # Rename wheel for Manylinux 2_28
-    if [[ $PLATFORM == "manylinux_2_28_x86_64" && $GPU_ARCH_TYPE != "cpu-s390x" && $GPU_ARCH_TYPE != "xpu" ]]; then
-        pkg_name=$(echo $(basename $pkg) | sed -e s#linux_x86_64#"${PLATFORM}"#)
-        zip -rq $pkg_name $PREIX*
-        rm -f $pkg
-        mv $pkg_name $(dirname $pkg)/$pkg_name
-    else
-        # zip up the wheel back
-        zip -rq $(basename $pkg) $PREIX*
-        # remove original wheel
-        rm -f $pkg
-        mv $(basename $pkg) $pkg
-    fi
+    # zip up the wheel back
+    zip -rq $(basename $pkg) $PREIX*
 
+    # replace original wheel
+    rm -f $pkg
+    mv $(basename $pkg) $pkg
     cd ..
     rm -rf tmp
 done
@@ -490,9 +474,9 @@ if [[ -z "$BUILD_PYTHONLESS" ]]; then
   echo "$(date) :: Running tests"
   pushd "$PYTORCH_ROOT"
 
-
+  #TODO: run_tests.sh and check_binary.sh should be moved to pytorch/pytorch project
   LD_LIBRARY_PATH=/usr/local/nvidia/lib64 \
-          "${PYTORCH_ROOT}/.ci/pytorch/run_tests.sh" manywheel "${py_majmin}" "$DESIRED_CUDA"
+          "/builder/run_tests.sh" manywheel "${py_majmin}" "$DESIRED_CUDA"
   popd
   echo "$(date) :: Finished tests"
 fi

.ci/manywheel/build_cuda.sh

@@ -14,7 +14,6 @@ export USE_CUDA_STATIC_LINK=1
 export INSTALL_TEST=0 # dont install test binaries into site-packages
 export USE_CUPTI_SO=0
 export USE_CUSPARSELT=${USE_CUSPARSELT:-1} # Enable if not disabled by libtorch build
-export USE_CUFILE=${USE_CUFILE:-1}
 
 # Keep an array of cmake variables to add to
 if [[ -z "$CMAKE_ARGS" ]]; then
@@ -44,6 +43,13 @@ if [[ -n "$DESIRED_CUDA" ]]; then
         fi
     fi
     echo "Using CUDA $CUDA_VERSION as determined by DESIRED_CUDA"
+
+    # There really has to be a better way to do this - eli
+    # Possibly limiting builds to specific cuda versions be delimiting images would be a choice
+    if [[ "$OS_NAME" == *"Ubuntu"* ]]; then
+        echo "Switching to CUDA version ${DESIRED_CUDA}"
+        /builder/conda/switch_cuda_version.sh "${DESIRED_CUDA}"
+    fi
 else
     CUDA_VERSION=$(nvcc --version|grep release|cut -f5 -d" "|cut -f1 -d",")
     echo "CUDA $CUDA_VERSION Detected"
@@ -53,15 +59,23 @@ cuda_version_nodot=$(echo $CUDA_VERSION | tr -d '.')
 
 TORCH_CUDA_ARCH_LIST="5.0;6.0;7.0;7.5;8.0;8.6"
 case ${CUDA_VERSION} in
-    12.8)
-        TORCH_CUDA_ARCH_LIST="${TORCH_CUDA_ARCH_LIST};9.0;10.0;12.0+PTX" #Ripping out 5.0 and 6.0 due to ld error
-        EXTRA_CAFFE2_CMAKE_FLAGS+=("-DATEN_NO_TEST=ON")
-        ;;
     12.6)
-        TORCH_CUDA_ARCH_LIST="${TORCH_CUDA_ARCH_LIST};9.0+PTX"
+        if [[ "$GPU_ARCH_TYPE" = "cuda-aarch64" ]]; then
+            TORCH_CUDA_ARCH_LIST="9.0"
+        else
+            TORCH_CUDA_ARCH_LIST="${TORCH_CUDA_ARCH_LIST};9.0+PTX"
+        fi
         EXTRA_CAFFE2_CMAKE_FLAGS+=("-DATEN_NO_TEST=ON")
         ;;
     12.4)
+        if [[ "$GPU_ARCH_TYPE" = "cuda-aarch64" ]]; then
+            TORCH_CUDA_ARCH_LIST="9.0"
+        else
+            TORCH_CUDA_ARCH_LIST="${TORCH_CUDA_ARCH_LIST};9.0"
+        fi
+        EXTRA_CAFFE2_CMAKE_FLAGS+=("-DATEN_NO_TEST=ON")
+        ;;
+    12.1)
         TORCH_CUDA_ARCH_LIST="${TORCH_CUDA_ARCH_LIST};9.0"
         EXTRA_CAFFE2_CMAKE_FLAGS+=("-DATEN_NO_TEST=ON")
         ;;
@@ -119,16 +133,7 @@ if [[ $USE_CUSPARSELT == "1" && $CUDA_VERSION == "11.8" ]]; then
         )
 fi
 
-
-# Turn USE_CUFILE off for CUDA 11.8, 12.4 since nvidia-cufile-cu11 and 1.9.0.20 are
-# not available in PYPI
-if [[ $CUDA_VERSION == "11.8" || $CUDA_VERSION == "12.4" ]]; then
-    export USE_CUFILE=0
-fi
-
-
-# CUDA_VERSION 12.4, 12.6, 12.8
-if [[ $CUDA_VERSION == 12* ]]; then
+if [[ $CUDA_VERSION == "12.4" || $CUDA_VERSION == "12.6" ]]; then
     export USE_STATIC_CUDNN=0
     # Try parallelizing nvcc as well
     export TORCH_NVCC_FLAGS="-Xfatbin -compress-all --threads 2"
@@ -169,16 +174,6 @@ if [[ $CUDA_VERSION == 12* ]]; then
             "libnvrtc.so.12"
             "libnvrtc-builtins.so"
         )
-        if [[ $USE_CUFILE == 1 ]]; then
-            DEPS_LIST+=(
-                "/usr/local/cuda/lib64/libcufile.so.0"
-                "/usr/local/cuda/lib64/libcufile_rdma.so.1"
-            )
-            DEPS_SONAME+=(
-                "libcufile.so.0"
-                "libcufile_rdma.so.1"
-            )
-        fi
     else
         echo "Using nvidia libs from pypi."
         CUDA_RPATHS=(
@@ -195,11 +190,6 @@ if [[ $CUDA_VERSION == 12* ]]; then
             '$ORIGIN/../../nvidia/nccl/lib'
             '$ORIGIN/../../nvidia/nvtx/lib'
         )
-        if [[ $USE_CUFILE == 1 ]]; then
-            CUDA_RPATHS+=(
-                '$ORIGIN/../../nvidia/cufile/lib'
-            )
-        fi
         CUDA_RPATHS=$(IFS=: ; echo "${CUDA_RPATHS[*]}")
         export C_SO_RPATH=$CUDA_RPATHS':$ORIGIN:$ORIGIN/lib'
         export LIB_SO_RPATH=$CUDA_RPATHS':$ORIGIN'
@@ -285,7 +275,7 @@ else
     exit 1
 fi
 
-# run_tests.sh requires DESIRED_CUDA to know what tests to exclude
+# builder/test.sh requires DESIRED_CUDA to know what tests to exclude
 export DESIRED_CUDA="$cuda_version_nodot"
 
 # Switch `/usr/local/cuda` to the desired CUDA version

.ci/manywheel/build_libtorch.sh

@@ -225,11 +225,11 @@ make_wheel_record() {
     FPATH=$1
     if echo $FPATH | grep RECORD >/dev/null 2>&1; then
         # if the RECORD file, then
-        echo "\"$FPATH\",,"
+        echo "$FPATH,,"
     else
         HASH=$(openssl dgst -sha256 -binary $FPATH | openssl base64 | sed -e 's/+/-/g' | sed -e 's/\//_/g' | sed -e 's/=//g')
         FSIZE=$(ls -nl $FPATH | awk '{print $5}')
-        echo "\"$FPATH\",sha256=$HASH,$FSIZE"
+        echo "$FPATH,sha256=$HASH,$FSIZE"
     fi
 }
 

.ci/manywheel/build_rocm.sh

@@ -107,29 +107,17 @@ if [[ $ROCM_INT -ge 60200 ]]; then
 fi
 
 OS_NAME=`awk -F= '/^NAME/{print $2}' /etc/os-release`
-if [[ "$OS_NAME" == *"CentOS Linux"* || "$OS_NAME" == *"AlmaLinux"* ]]; then
+if [[ "$OS_NAME" == *"CentOS Linux"* ]]; then
     LIBGOMP_PATH="/usr/lib64/libgomp.so.1"
     LIBNUMA_PATH="/usr/lib64/libnuma.so.1"
     LIBELF_PATH="/usr/lib64/libelf.so.1"
-    if [[ "$OS_NAME" == *"CentOS Linux"* ]]; then
-        LIBTINFO_PATH="/usr/lib64/libtinfo.so.5"
-    else
-        LIBTINFO_PATH="/usr/lib64/libtinfo.so.6"
-    fi
+    LIBTINFO_PATH="/usr/lib64/libtinfo.so.5"
     LIBDRM_PATH="/opt/amdgpu/lib64/libdrm.so.2"
     LIBDRM_AMDGPU_PATH="/opt/amdgpu/lib64/libdrm_amdgpu.so.1"
-    if [[ $ROCM_INT -ge 60100 && $ROCM_INT -lt 60300 ]]; then
+    if [[ $ROCM_INT -ge 60100 ]]; then
         # Below libs are direct dependencies of libhipsolver
         LIBSUITESPARSE_CONFIG_PATH="/lib64/libsuitesparseconfig.so.4"
-        if [[ "$OS_NAME" == *"CentOS Linux"* ]]; then
-            LIBCHOLMOD_PATH="/lib64/libcholmod.so.2"
-            # Below libs are direct dependencies of libsatlas
-            LIBGFORTRAN_PATH="/lib64/libgfortran.so.3"
-        else
-            LIBCHOLMOD_PATH="/lib64/libcholmod.so.3"
-            # Below libs are direct dependencies of libsatlas
-            LIBGFORTRAN_PATH="/lib64/libgfortran.so.5"
-        fi
+        LIBCHOLMOD_PATH="/lib64/libcholmod.so.2"
         # Below libs are direct dependencies of libcholmod
         LIBAMD_PATH="/lib64/libamd.so.2"
         LIBCAMD_PATH="/lib64/libcamd.so.2"
@@ -137,6 +125,7 @@ if [[ "$OS_NAME" == *"CentOS Linux"* || "$OS_NAME" == *"AlmaLinux"* ]]; then
         LIBCOLAMD_PATH="/lib64/libcolamd.so.2"
         LIBSATLAS_PATH="/lib64/atlas/libsatlas.so.3"
         # Below libs are direct dependencies of libsatlas
+        LIBGFORTRAN_PATH="/lib64/libgfortran.so.3"
         LIBQUADMATH_PATH="/lib64/libquadmath.so.0"
     fi
     MAYBE_LIB64=lib64
@@ -151,7 +140,7 @@ elif [[ "$OS_NAME" == *"Ubuntu"* ]]; then
     fi
     LIBDRM_PATH="/usr/lib/x86_64-linux-gnu/libdrm.so.2"
     LIBDRM_AMDGPU_PATH="/usr/lib/x86_64-linux-gnu/libdrm_amdgpu.so.1"
-    if [[ $ROCM_INT -ge 60100 && $ROCM_INT -lt 60300 ]]; then
+    if [[ $ROCM_INT -ge 60100 ]]; then
         # Below libs are direct dependencies of libhipsolver
         LIBCHOLMOD_PATH="/lib/x86_64-linux-gnu/libcholmod.so.3"
         # Below libs are direct dependencies of libcholmod
@@ -186,6 +175,12 @@ do
     OS_SO_FILES[${#OS_SO_FILES[@]}]=$file_name # Append lib to array
 done
 
+# PyTorch-version specific
+# AOTriton dependency only for PyTorch >= 2.4
+if (( $(echo "${PYTORCH_VERSION} 2.4" | awk '{print ($1 >= $2)}') )); then
+    ROCM_SO_FILES+=("libaotriton_v2.so")
+fi
+
 # rocBLAS library files
 ROCBLAS_LIB_SRC=$ROCM_HOME/lib/rocblas/library
 ROCBLAS_LIB_DST=lib/rocblas/library

.ci/pytorch/build.sh

@@ -1,6 +1,6 @@
 #!/bin/bash
 
-set -ex -o pipefail
+set -ex
 
 # Required environment variable: $BUILD_ENVIRONMENT
 # (This is set by default in the Docker images we build, so you don't
@@ -87,7 +87,7 @@ else
 
   # Workaround required for MKL library linkage
   # https://github.com/pytorch/pytorch/issues/119557
-  if [[ "$ANACONDA_PYTHON_VERSION" = "3.12" || "$ANACONDA_PYTHON_VERSION" = "3.13" ]]; then
+  if [ "$ANACONDA_PYTHON_VERSION" = "3.12" ]; then
     export CMAKE_LIBRARY_PATH="/opt/conda/envs/py_$ANACONDA_PYTHON_VERSION/lib/"
     export CMAKE_INCLUDE_PATH="/opt/conda/envs/py_$ANACONDA_PYTHON_VERSION/include/"
   fi
@@ -191,7 +191,7 @@ fi
 
 # We only build FlashAttention files for CUDA 8.0+, and they require large amounts of
 # memory to build and will OOM
-if [[ "$BUILD_ENVIRONMENT" == *cuda* ]] && [[ 1 -eq $(echo "${TORCH_CUDA_ARCH_LIST} >= 8.0" | bc) ]]; then
+if [[ "$BUILD_ENVIRONMENT" == *cuda* ]] && [[ "$TORCH_CUDA_ARCH_LIST" == *"8.6"* || "$TORCH_CUDA_ARCH_LIST" == *"8.0"* ]]; then
   echo "WARNING: FlashAttention files require large amounts of memory to build and will OOM"
   echo "Setting MAX_JOBS=(nproc-2)/3 to reduce memory usage"
   export MAX_JOBS="$(( $(nproc --ignore=2) / 3 ))"
@@ -228,9 +228,9 @@ if [[ "$BUILD_ENVIRONMENT" == *-debug* ]]; then
   export CMAKE_BUILD_TYPE=RelWithAssert
 fi
 
-# Do not change workspace permissions for ROCm and s390x CI jobs
+# Do not change workspace permissions for ROCm CI jobs
 # as it can leave workspace with bad permissions for cancelled jobs
-if [[ "$BUILD_ENVIRONMENT" != *rocm* && "$BUILD_ENVIRONMENT" != *s390x* && -d /var/lib/jenkins/workspace ]]; then
+if [[ "$BUILD_ENVIRONMENT" != *rocm* && "$BUILD_ENVIRONMENT" != *s390x* ]]; then
   # Workaround for dind-rootless userid mapping (https://github.com/pytorch/ci-infra/issues/96)
   WORKSPACE_ORIGINAL_OWNER_ID=$(stat -c '%u' "/var/lib/jenkins/workspace")
   cleanup_workspace() {
@@ -247,9 +247,10 @@ if [[ "$BUILD_ENVIRONMENT" != *rocm* && "$BUILD_ENVIRONMENT" != *s390x* && -d /v
 fi
 
 if [[ "$BUILD_ENVIRONMENT" == *-bazel-* ]]; then
-  set -e -o pipefail
+  set -e
 
   get_bazel
+  install_sccache_nvcc_for_bazel
 
   # Leave 1 CPU free and use only up to 80% of memory to reduce the change of crashing
   # the runner
@@ -278,7 +279,7 @@ else
           "$BUILD_ENVIRONMENT" != *xla* ]]; then
       if [[ "$BUILD_ENVIRONMENT" != *py3.8* ]]; then
         # Install numpy-2.0.2 for builds which are backward compatible with 1.X
-        python -mpip install numpy==2.0.2
+        python -mpip install --pre numpy==2.0.2
       fi
 
       WERROR=1 python setup.py clean
@@ -395,7 +396,7 @@ if [[ "$BUILD_ENVIRONMENT" != *libtorch* && "$BUILD_ENVIRONMENT" != *bazel* ]];
   # don't do this for libtorch as libtorch is C++ only and thus won't have python tests run on its build
   python tools/stats/export_test_times.py
 fi
-# don't do this for bazel or s390x as they don't use sccache
-if [[ "$BUILD_ENVIRONMENT" != *s390x* && "$BUILD_ENVIRONMENT" != *-bazel-* ]]; then
+
+if [[ "$BUILD_ENVIRONMENT" != *s390x* ]]; then
   print_sccache_stats
 fi

.ci/pytorch/check_binary.sh

@@ -1,394 +0,0 @@
-#!/bin/bash
-
-# shellcheck disable=SC2086,SC2006,SC2207,SC2076,SC2155,SC2046,SC1091,SC2143
-# TODO: Re-enable shellchecks above
-
-set -eux -o pipefail
-
-# This script checks the following things on binaries
-# 1. The gcc abi matches DESIRED_DEVTOOLSET
-# 2. MacOS binaries do not link against OpenBLAS
-# 3. There are no protobuf symbols of any sort anywhere (turned off, because
-#    this is currently not true)
-# 4. Standard Python imports work
-# 5. MKL is available everywhere except for MacOS wheels
-# 6. XNNPACK is available everywhere except for MacOS wheels
-# 7. CUDA is setup correctly and does not hang
-# 8. Magma is available for CUDA builds
-# 9. CuDNN is available for CUDA builds
-#
-# This script needs the env variables DESIRED_PYTHON, DESIRED_CUDA,
-# DESIRED_DEVTOOLSET and PACKAGE_TYPE
-#
-# This script expects PyTorch to be installed into the active Python (the
-# Python returned by `which python`). Or, if this is testing a libtorch
-# Pythonless binary, then it expects to be in the root folder of the unzipped
-# libtorch package.
-
-
-if [[ -z ${DESIRED_PYTHON:-} ]]; then
-  export DESIRED_PYTHON=${MATRIX_PYTHON_VERSION:-}
-fi
-if [[ -z ${DESIRED_CUDA:-} ]]; then
-  export DESIRED_CUDA=${MATRIX_DESIRED_CUDA:-}
-fi
-if [[ -z ${DESIRED_DEVTOOLSET:-} ]]; then
-  export DESIRED_DEVTOOLSET=${MATRIX_DESIRED_DEVTOOLSET:-}
-fi
-if [[ -z ${PACKAGE_TYPE:-} ]]; then
-  export PACKAGE_TYPE=${MATRIX_PACKAGE_TYPE:-}
-fi
-
-# The install root depends on both the package type and the os
-# All MacOS packages use conda, even for the wheel packages.
-if [[ "$PACKAGE_TYPE" == libtorch ]]; then
-  # NOTE: Only $PWD works on both CentOS and Ubuntu
-  export install_root="$PWD"
-else
-
-  if [[ $DESIRED_PYTHON =~ ([0-9].[0-9]+)t ]]; then
-    # For python that is maj.mint keep original version
-    py_dot="$DESIRED_PYTHON"
-  elif [[ $DESIRED_PYTHON =~ ([0-9].[0-9]+) ]];  then
-    # Strip everything but major.minor from DESIRED_PYTHON version
-    py_dot="${BASH_REMATCH[0]}"
-  else
-    echo "Unexpected ${DESIRED_PYTHON} format"
-    exit 1
-  fi
-  export install_root="$(dirname $(which python))/../lib/python${py_dot}/site-packages/torch/"
-fi
-
-###############################################################################
-# Setup XPU ENV
-###############################################################################
-if [[ "$DESIRED_CUDA" == 'xpu' ]]; then
-  set +u
-  # Refer https://www.intel.com/content/www/us/en/developer/articles/tool/pytorch-prerequisites-for-intel-gpus.html
-  source /opt/intel/oneapi/compiler/latest/env/vars.sh
-  source /opt/intel/oneapi/pti/latest/env/vars.sh
-fi
-
-###############################################################################
-# Check GCC ABI
-###############################################################################
-
-# NOTE [ Building libtorch with old vs. new gcc ABI ]
-#
-# Packages built with one version of ABI could not be linked against by client
-# C++ libraries that were compiled using the other version of ABI. Since both
-# gcc ABIs are still common in the wild, we need to support both ABIs. Currently:
-#
-# - All the nightlies built on CentOS 7 + devtoolset7 use the old gcc ABI.
-# - All the nightlies built on Ubuntu 16.04 + gcc 5.4 use the new gcc ABI.
-
-echo "Checking that the gcc ABI is what we expect"
-if [[ "$(uname)" != 'Darwin' ]]; then
-  function is_expected() {
-    if [[ "$DESIRED_DEVTOOLSET" == *"cxx11-abi"* || "$DESIRED_CUDA" == *"rocm"* ]]; then
-      if [[ "$1" -gt 0 || "$1" == "ON " ]]; then
-        echo 1
-      fi
-    else
-      if [[ -z "$1" || "$1" == 0 || "$1" == "OFF" ]]; then
-        echo 1
-      fi
-    fi
-  }
-
-  # First we check that the env var in TorchConfig.cmake is correct
-
-  # We search for D_GLIBCXX_USE_CXX11_ABI=1 in torch/TorchConfig.cmake
-  torch_config="${install_root}/share/cmake/Torch/TorchConfig.cmake"
-  if [[ ! -f "$torch_config" ]]; then
-    echo "No TorchConfig.cmake found!"
-    ls -lah "$install_root/share/cmake/Torch"
-    exit 1
-  fi
-  echo "Checking the TorchConfig.cmake"
-  cat "$torch_config"
-
-  # The sed call below is
-  #   don't print lines by default (only print the line we want)
-  # -n
-  #   execute the following expression
-  # e
-  #   replace lines that match with the first capture group and print
-  # s/.*D_GLIBCXX_USE_CXX11_ABI=\(.\)".*/\1/p
-  #   any characters, D_GLIBCXX_USE_CXX11_ABI=, exactly one any character, a
-  #   quote, any characters
-  #   Note the exactly one single character after the '='. In the case that the
-  #     variable is not set the '=' will be followed by a '"' immediately and the
-  #     line will fail the match and nothing will be printed; this is what we
-  #     want.  Otherwise it will capture the 0 or 1 after the '='.
-  # /.*D_GLIBCXX_USE_CXX11_ABI=\(.\)".*/
-  #   replace the matched line with the capture group and print
-  # /\1/p
-  actual_gcc_abi="$(sed -ne 's/.*D_GLIBCXX_USE_CXX11_ABI=\(.\)".*/\1/p' < "$torch_config")"
-  if [[ "$(is_expected "$actual_gcc_abi")" != 1 ]]; then
-    echo "gcc ABI $actual_gcc_abi not as expected."
-    exit 1
-  fi
-
-  # We also check that there are [not] cxx11 symbols in libtorch
-  #
-  echo "Checking that symbols in libtorch.so have the right gcc abi"
-  python3 "$(dirname ${BASH_SOURCE[0]})/smoke_test/check_binary_symbols.py"
-
-  echo "cxx11 symbols seem to be in order"
-fi # if on Darwin
-
-###############################################################################
-# Check for no OpenBLAS
-# TODO Check for no Protobuf symbols (not finished)
-# Print *all* runtime dependencies
-###############################################################################
-# We have to loop through all shared libraries for this
-if [[ "$(uname)" == 'Darwin' ]]; then
-  all_dylibs=($(find "$install_root" -name '*.dylib'))
-  for dylib in "${all_dylibs[@]}"; do
-    echo "All dependencies of $dylib are $(otool -L $dylib) with rpath $(otool -l $dylib | grep LC_RPATH -A2)"
-
-    # Check that OpenBlas is not linked to on Macs
-    echo "Checking the OpenBLAS is not linked to"
-    if [[ -n "$(otool -L $dylib | grep -i openblas)" ]]; then
-      echo "ERROR: Found openblas as a dependency of $dylib"
-      echo "Full dependencies is: $(otool -L $dylib)"
-      exit 1
-    fi
-
-    # Check for protobuf symbols
-    #proto_symbols="$(nm $dylib | grep protobuf)" || true
-    #if [[ -n "$proto_symbols" ]]; then
-    #  echo "ERROR: Detected protobuf symbols in $dylib"
-    #  echo "Symbols are $proto_symbols"
-    #  exit 1
-    #fi
-  done
-else
-  all_libs=($(find "$install_root" -name '*.so'))
-  for lib in "${all_libs[@]}"; do
-    echo "All dependencies of $lib are $(ldd $lib) with runpath $(objdump -p $lib | grep RUNPATH)"
-
-    # Check for protobuf symbols
-    #proto_symbols=$(nm $lib | grep protobuf) || true
-    #if [[ -n "$proto_symbols" ]]; then
-    #  echo "ERROR: Detected protobuf symbols in $lib"
-    #  echo "Symbols are $proto_symbols"
-    #  exit 1
-    #fi
-  done
-fi
-
-setup_link_flags () {
-  REF_LIB="-Wl,-R${install_root}/lib"
-  if [[ "$(uname)" == 'Darwin' ]]; then
-    REF_LIB="-Wl,-rpath ${install_root}/lib"
-  fi
-  ADDITIONAL_LINKER_FLAGS=""
-  if [[ "$(uname)" == 'Linux' ]]; then
-    ADDITIONAL_LINKER_FLAGS="-Wl,--no-as-needed"
-  fi
-  C10_LINK_FLAGS=""
-  if [ -f "${install_root}/lib/libc10.so" ] || [ -f "${install_root}/lib/libc10.dylib" ]; then
-    C10_LINK_FLAGS="-lc10"
-  fi
-  TORCH_CPU_LINK_FLAGS=""
-  if [ -f "${install_root}/lib/libtorch_cpu.so" ] || [ -f "${install_root}/lib/libtorch_cpu.dylib" ]; then
-    TORCH_CPU_LINK_FLAGS="-ltorch_cpu"
-  fi
-  TORCH_CUDA_LINK_FLAGS=""
-  if [ -f "${install_root}/lib/libtorch_cuda.so" ] || [ -f "${install_root}/lib/libtorch_cuda.dylib" ]; then
-    TORCH_CUDA_LINK_FLAGS="-ltorch_cuda"
-  elif [ -f "${install_root}/lib/libtorch_cuda_cpp.so" ] && [ -f "${install_root}/lib/libtorch_cuda_cpp.so" ] || \
-    [ -f "${install_root}/lib/libtorch_cuda_cu.dylib" ] && [ -f "${install_root}/lib/libtorch_cuda_cu.dylib" ]; then
-    TORCH_CUDA_LINK_FLAGS="-ltorch_cuda_cpp -ltorch_cuda_cu"
-  fi
-}
-
-TEST_CODE_DIR="$(dirname $(realpath ${BASH_SOURCE[0]}))/test_example_code"
-build_and_run_example_cpp () {
-  if [[ "$DESIRED_DEVTOOLSET" == *"cxx11-abi"* ]]; then
-    GLIBCXX_USE_CXX11_ABI=1
-  else
-    GLIBCXX_USE_CXX11_ABI=0
-  fi
-  setup_link_flags
-  g++ ${TEST_CODE_DIR}/$1.cpp -I${install_root}/include -I${install_root}/include/torch/csrc/api/include -D_GLIBCXX_USE_CXX11_ABI=$GLIBCXX_USE_CXX11_ABI -std=gnu++17 -L${install_root}/lib ${REF_LIB} ${ADDITIONAL_LINKER_FLAGS} -ltorch $TORCH_CPU_LINK_FLAGS $TORCH_CUDA_LINK_FLAGS $C10_LINK_FLAGS -o $1
-  ./$1
-}
-
-build_example_cpp_with_incorrect_abi () {
-  if [[ "$DESIRED_DEVTOOLSET" == *"cxx11-abi"* ]]; then
-    GLIBCXX_USE_CXX11_ABI=0
-  else
-    GLIBCXX_USE_CXX11_ABI=1
-  fi
-  set +e
-  setup_link_flags
-  g++ ${TEST_CODE_DIR}/$1.cpp -I${install_root}/include -I${install_root}/include/torch/csrc/api/include -D_GLIBCXX_USE_CXX11_ABI=$GLIBCXX_USE_CXX11_ABI -std=gnu++17 -L${install_root}/lib ${REF_LIB} ${ADDITIONAL_LINKER_FLAGS} -ltorch $TORCH_CPU_LINK_FLAGS $TORCH_CUDA_LINK_FLAGS $C10_LINK_FLAGS -o $1
-  ERRCODE=$?
-  set -e
-  if [ "$ERRCODE" -eq "0" ]; then
-    echo "Building example with incorrect ABI didn't throw error. Aborting."
-    exit 1
-  else
-    echo "Building example with incorrect ABI throws expected error. Proceeding."
-  fi
-}
-
-###############################################################################
-# Check simple Python/C++ calls
-###############################################################################
-if [[ "$PACKAGE_TYPE" == 'libtorch' ]]; then
-  # NS: Set LD_LIBRARY_PATH for CUDA builds, but perhaps it should be removed
-  if [[ "$DESIRED_CUDA" == "cu"* ]]; then
-    export LD_LIBRARY_PATH=/usr/local/cuda/lib64
-  fi
-  build_and_run_example_cpp simple-torch-test
-  # `_GLIBCXX_USE_CXX11_ABI` is always ignored by gcc in devtoolset7, so we test
-  # the expected failure case for Ubuntu 16.04 + gcc 5.4 only.
-  if [[ "$DESIRED_DEVTOOLSET" == *"cxx11-abi"* ]]; then
-    build_example_cpp_with_incorrect_abi simple-torch-test
-  fi
-else
-  pushd /tmp
-  python -c 'import torch'
-  popd
-fi
-
-###############################################################################
-# Check torch.git_version
-###############################################################################
-if [[ "$PACKAGE_TYPE" != 'libtorch' ]]; then
-  pushd /tmp
-  python -c 'import torch; assert torch.version.git_version != "Unknown"'
-  python -c 'import torch; assert torch.version.git_version != None'
-  popd
-fi
-
-
-###############################################################################
-# Check for MKL
-###############################################################################
-
-if [[ "$PACKAGE_TYPE" == 'libtorch' ]]; then
-  echo "Checking that MKL is available"
-  build_and_run_example_cpp check-torch-mkl
-elif [[ "$(uname -m)" != "arm64" && "$(uname -m)" != "s390x" ]]; then
-  if [[ "$(uname)" != 'Darwin' || "$PACKAGE_TYPE" != *wheel ]]; then
-    if [[ "$(uname -m)" == "aarch64" ]]; then
-      echo "Checking that MKLDNN is available on aarch64"
-      pushd /tmp
-      python -c 'import torch; exit(0 if torch.backends.mkldnn.is_available() else 1)'
-      popd
-    else
-      echo "Checking that MKL is available"
-      pushd /tmp
-      python -c 'import torch; exit(0 if torch.backends.mkl.is_available() else 1)'
-      popd
-    fi
-  fi
-fi
-
-###############################################################################
-# Check for XNNPACK
-###############################################################################
-
-if [[ "$PACKAGE_TYPE" == 'libtorch' ]]; then
-  echo "Checking that XNNPACK is available"
-  build_and_run_example_cpp check-torch-xnnpack
-else
-  if [[ "$(uname)" != 'Darwin' || "$PACKAGE_TYPE" != *wheel ]] && [[ "$(uname -m)" != "s390x"  ]]; then
-    echo "Checking that XNNPACK is available"
-    pushd /tmp
-    python -c 'import torch.backends.xnnpack; exit(0 if torch.backends.xnnpack.enabled else 1)'
-    popd
-  fi
-fi
-
-###############################################################################
-# Check CUDA configured correctly
-###############################################################################
-# Skip these for Windows machines without GPUs
-if [[ "$OSTYPE" == "msys" ]]; then
-    GPUS=$(wmic path win32_VideoController get name)
-    if [[ ! "$GPUS" == *NVIDIA* ]]; then
-        echo "Skip CUDA tests for machines without a Nvidia GPU card"
-        exit 0
-    fi
-fi
-
-# Test that CUDA builds are setup correctly
-if [[ "$DESIRED_CUDA" != 'cpu' && "$DESIRED_CUDA" != 'xpu' && "$DESIRED_CUDA" != 'cpu-cxx11-abi' && "$DESIRED_CUDA" != *"rocm"* && "$(uname -m)" != "s390x" ]]; then
-  if [[ "$PACKAGE_TYPE" == 'libtorch' ]]; then
-    build_and_run_example_cpp check-torch-cuda
-  else
-    pushd /tmp
-    echo "Checking that CUDA archs are setup correctly"
-    timeout 20 python -c 'import torch; torch.randn([3,5]).cuda()'
-
-    # These have to run after CUDA is initialized
-
-    echo "Checking that magma is available"
-    python -c 'import torch; torch.rand(1).cuda(); exit(0 if torch.cuda.has_magma else 1)'
-
-    echo "Checking that CuDNN is available"
-    python -c 'import torch; exit(0 if torch.backends.cudnn.is_available() else 1)'
-
-    # Validates builds is free of linker regressions reported in https://github.com/pytorch/pytorch/issues/57744
-    echo "Checking that exception handling works"
-    python -c "import torch; from unittest import TestCase;TestCase().assertRaises(RuntimeError, lambda:torch.eye(7, 7, device='cuda:7'))"
-
-    echo "Checking that basic RNN works"
-    python ${TEST_CODE_DIR}/rnn_smoke.py
-
-    echo "Checking that basic CNN works"
-    python "${TEST_CODE_DIR}/cnn_smoke.py"
-
-    echo "Test that linalg works"
-    python -c "import torch;x=torch.rand(3,3,device='cuda');print(torch.linalg.svd(torch.mm(x.t(), x)))"
-
-    popd
-  fi # if libtorch
-fi # if cuda
-
-##########################
-# Run parts of smoke tests
-##########################
-if [[ "$PACKAGE_TYPE" != 'libtorch' ]]; then
-  pushd "$(dirname ${BASH_SOURCE[0]})/smoke_test"
-  python -c "from smoke_test import test_linalg; test_linalg()"
-  if [[ "$DESIRED_CUDA" == *cuda* ]]; then
-    python -c "from smoke_test import test_linalg; test_linalg('cuda')"
-  fi
-  popd
-fi
-
-###############################################################################
-# Check PyTorch supports TCP_TLS gloo transport
-###############################################################################
-
-if [[ "$(uname)" == 'Linux' && "$PACKAGE_TYPE" != 'libtorch' ]]; then
-  GLOO_CHECK="import torch.distributed as dist
-try:
-    dist.init_process_group('gloo', rank=0, world_size=1)
-except RuntimeError as e:
-    print(e)
-"
-  RESULT=`GLOO_DEVICE_TRANSPORT=TCP_TLS MASTER_ADDR=localhost MASTER_PORT=63945 python -c "$GLOO_CHECK"`
-  GLOO_TRANSPORT_IS_NOT_SUPPORTED='gloo transport is not supported'
-  if [[ "$RESULT" =~ "$GLOO_TRANSPORT_IS_NOT_SUPPORTED" ]]; then
-    echo "PyTorch doesn't support TLS_TCP transport, please build with USE_GLOO_WITH_OPENSSL=1"
-    exit 1
-  fi
-fi
-
-###############################################################################
-# Check for C++ ABI compatibility between gcc7 and gcc9 compiled binaries
-###############################################################################
-if [[ "$(uname)" == 'Linux' &&  "$PACKAGE_TYPE" == 'manywheel' ]]; then
-  pushd /tmp
-  python -c "import torch; exit(0 if torch.compiled_with_cxx11_abi() else (0 if torch._C._PYBIND11_BUILD_ABI == '_cxxabi1011' else 1))"
-  popd
-fi

.ci/pytorch/common.sh

@@ -3,7 +3,7 @@
 # Common setup for all Jenkins scripts
 # shellcheck source=./common_utils.sh
 source "$(dirname "${BASH_SOURCE[0]}")/common_utils.sh"
-set -ex -o pipefail
+set -ex
 
 # Required environment variables:
 #   $BUILD_ENVIRONMENT (should be set by your Docker image)

.ci/pytorch/common_utils.sh

@@ -111,6 +111,26 @@ function get_bazel() {
   chmod u+x tools/bazel
 }
 
+# This function is bazel specific because of the bug
+# in the bazel that requires some special paths massaging
+# as a workaround. See
+# https://github.com/bazelbuild/bazel/issues/10167
+function install_sccache_nvcc_for_bazel() {
+  sudo mv /usr/local/cuda/bin/nvcc /usr/local/cuda/bin/nvcc-real
+
+  # Write the `/usr/local/cuda/bin/nvcc`
+  cat << EOF | sudo tee /usr/local/cuda/bin/nvcc
+#!/bin/sh
+if [ \$(env -u LD_PRELOAD ps -p \$PPID -o comm=) != sccache ]; then
+  exec sccache /usr/local/cuda/bin/nvcc "\$@"
+else
+  exec external/local_cuda/cuda/bin/nvcc-real "\$@"
+fi
+EOF
+
+  sudo chmod +x /usr/local/cuda/bin/nvcc
+}
+
 function install_monkeytype {
   # Install MonkeyType
   pip_install MonkeyType
@@ -160,7 +180,7 @@ function install_torchvision() {
 }
 
 function install_tlparse() {
-  pip_install --user "tlparse==0.3.30"
+  pip_install --user "tlparse==0.3.25"
   PATH="$(python -m site --user-base)/bin:$PATH"
 }
 
@@ -169,34 +189,24 @@ function install_torchrec_and_fbgemm() {
   torchrec_commit=$(get_pinned_commit torchrec)
   local fbgemm_commit
   fbgemm_commit=$(get_pinned_commit fbgemm)
-  if [[ "$BUILD_ENVIRONMENT" == *rocm* ]] ; then
-    fbgemm_commit=$(get_pinned_commit fbgemm_rocm)
-  fi
   pip_uninstall torchrec-nightly
   pip_uninstall fbgemm-gpu-nightly
   pip_install setuptools-git-versioning scikit-build pyre-extensions
 
-  if [[ "$BUILD_ENVIRONMENT" == *rocm* ]] ; then
-    # install torchrec first because it installs fbgemm nightly on top of rocm fbgemm
-    pip_install --no-use-pep517 --user "git+https://github.com/pytorch/torchrec.git@${torchrec_commit}"
-    pip_uninstall fbgemm-gpu-nightly
+  # TODO (huydhn): I still have no clue on why sccache doesn't work with only fbgemm_gpu here, but it
+  # seems to be an sccache-related issue
+  if [[ "$IS_A100_RUNNER" == "1" ]]; then
+    unset CMAKE_CUDA_COMPILER_LAUNCHER
+    sudo mv /opt/cache/bin /opt/cache/bin-backup
+  fi
 
-    pip_install tabulate  # needed for newer fbgemm
-    pip_install patchelf  # needed for rocm fbgemm
-    git clone --recursive https://github.com/pytorch/fbgemm
-    pushd fbgemm/fbgemm_gpu
-    git checkout "${fbgemm_commit}"
-    python setup.py install \
-      --package_variant=rocm \
-      -DHIP_ROOT_DIR="${ROCM_PATH}" \
-      -DCMAKE_C_FLAGS="-DTORCH_USE_HIP_DSA" \
-      -DCMAKE_CXX_FLAGS="-DTORCH_USE_HIP_DSA"
-    popd
-    rm -rf fbgemm
-  else
-    # See https://github.com/pytorch/pytorch/issues/106971
-    CUDA_PATH=/usr/local/cuda-12.1 pip_install --no-use-pep517 --user "git+https://github.com/pytorch/FBGEMM.git@${fbgemm_commit}#egg=fbgemm-gpu&subdirectory=fbgemm_gpu"
-    pip_install --no-use-pep517 --user "git+https://github.com/pytorch/torchrec.git@${torchrec_commit}"
+  # See https://github.com/pytorch/pytorch/issues/106971
+  CUDA_PATH=/usr/local/cuda-12.1 pip_install --no-use-pep517 --user "git+https://github.com/pytorch/FBGEMM.git@${fbgemm_commit}#egg=fbgemm-gpu&subdirectory=fbgemm_gpu"
+  pip_install --no-use-pep517 --user "git+https://github.com/pytorch/torchrec.git@${torchrec_commit}"
+
+  if [[ "$IS_A100_RUNNER" == "1" ]]; then
+    export CMAKE_CUDA_COMPILER_LAUNCHER=/opt/cache/bin/sccache
+    sudo mv /opt/cache/bin-backup /opt/cache/bin
   fi
 }
 
@@ -226,11 +236,6 @@ function checkout_install_torchbench() {
     # to install and test other models
     python install.py --continue_on_fail
   fi
-
-  # TODO (huydhn): transformers-4.44.2 added by https://github.com/pytorch/benchmark/pull/2488
-  # is regressing speedup metric. This needs to be investigated further
-  pip install transformers==4.38.1
-
   echo "Print all dependencies after TorchBench is installed"
   python -mpip freeze
   popd

.ci/pytorch/cpp_doc_push_script.sh

@@ -40,7 +40,7 @@ echo "Building PyTorch C++ API docs..."
 rm -rf cppdocs
 git clone https://github.com/pytorch/cppdocs
 
-set -ex -o pipefail
+set -ex
 
 # Generate ATen files
 pushd "${pt_checkout}"

.ci/pytorch/functorch_doc_push_script.sh

@@ -5,7 +5,7 @@ pt_checkout="/var/lib/jenkins/workspace"
 source "$pt_checkout/.ci/pytorch/common_utils.sh"
 echo "functorch_doc_push_script.sh: Invoked with $*"
 
-set -ex -o pipefail
+set -ex
 
 version=${DOCS_VERSION:-nightly}
 echo "version: $version"

.ci/pytorch/install_cache_xla.sh

@@ -6,7 +6,7 @@
 # return the same thing, ex checks for for rocm, CUDA, and changing the path
 # where sccache is installed, and not changing /etc/environment.
 
-set -ex -o pipefail
+set -ex
 
 install_binary() {
   echo "Downloading sccache binary from S3 repo"

.ci/pytorch/macos-test.sh

@@ -18,9 +18,6 @@ if [[ ! $(python -c "import torch; print(int(torch.backends.openmp.is_available(
 fi
 popd
 
-# enable debug asserts in serialization
-export TORCH_SERIALIZATION_DEBUG=1
-
 setup_test_python() {
   # The CircleCI worker hostname doesn't resolve to an address.
   # This environment variable makes ProcessGroupGloo default to
@@ -193,19 +190,11 @@ test_torchbench_perf() {
   TEST_REPORTS_DIR=$(pwd)/test/test-reports
   mkdir -p "$TEST_REPORTS_DIR"
 
-  local backend=eager
-  local dtype=notset
-  local device=mps
-
   echo "Setup complete, launching torchbench training performance run"
-  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py \
-    --performance --backend "$backend" --training --devices "$device" \
-    --output "$TEST_REPORTS_DIR/inductor_${backend}_torchbench_${dtype}_training_${device}_performance.csv"
+  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py --performance --backend eager --training --devices mps --output "$TEST_REPORTS_DIR/torchbench_training.csv"
 
   echo "Launching torchbench inference performance run"
-  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py \
-    --performance --backend "$backend" --inference --devices "$device" \
-    --output "$TEST_REPORTS_DIR/inductor_${backend}_torchbench_${dtype}_inference_${device}_performance.csv"
+  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py --performance --backend eager --inference --devices mps --output "$TEST_REPORTS_DIR/torchbench_training.csv"
 
   echo "Pytorch benchmark on mps device completed"
 }
@@ -220,27 +209,26 @@ test_torchbench_smoketest() {
 
   TEST_REPORTS_DIR=$(pwd)/test/test-reports
   mkdir -p "$TEST_REPORTS_DIR"
-
-  local backend=eager
-  local dtype=notset
-  local device=mps
-
-  touch "$TEST_REPORTS_DIR/inductor_${backend}_torchbench_${dtype}_training_${device}_performance.csv"
-  touch "$TEST_REPORTS_DIR/inductor_${backend}_torchbench_${dtype}_inference_${device}_performance.csv"
+  touch "$TEST_REPORTS_DIR"/torchbench_training.csv
+  touch "$TEST_REPORTS_DIR"/torchbench_inference.csv
 
   echo "Setup complete, launching torchbench training performance run"
-  for model in hf_T5 llama BERT_pytorch dcgan hf_GPT2 yolov3 resnet152; do
-    PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py \
-      --performance --only "$model" --backend "$backend" --training --devices "$device" \
-      --output "$TEST_REPORTS_DIR/inductor_${backend}_torchbench_${dtype}_training_${device}_performance.csv"
-  done
+  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py --performance --only hf_T5 --backend eager --training --devices mps --output "$TEST_REPORTS_DIR/torchbench_training.csv"
+  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py --performance --only llama --backend eager --training --devices mps --output "$TEST_REPORTS_DIR/torchbench_training.csv"
+  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py --performance --only BERT_pytorch --backend eager --training --devices mps --output "$TEST_REPORTS_DIR/torchbench_training.csv"
+  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py --performance --only dcgan --backend eager --training --devices mps --output "$TEST_REPORTS_DIR/torchbench_training.csv"
+  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py --performance --only hf_GPT2 --backend eager --training --devices mps --output "$TEST_REPORTS_DIR/torchbench_training.csv"
+  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py --performance --only yolov3 --backend eager --training --devices mps --output "$TEST_REPORTS_DIR/torchbench_training.csv"
+  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py --performance --only resnet152 --backend eager --training --devices mps --output "$TEST_REPORTS_DIR/torchbench_training.csv"
 
   echo "Launching torchbench inference performance run"
-  for model in hf_T5 llama BERT_pytorch dcgan hf_GPT2 yolov3 resnet152; do
-    PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py \
-      --performance --only "$model" --backend "$backend" --inference --devices "$device" \
-      --output "$TEST_REPORTS_DIR/inductor_${backend}_torchbench_${dtype}_inference_${device}_performance.csv"
-  done
+  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py --performance --only hf_T5 --backend eager --inference --devices mps --output "$TEST_REPORTS_DIR/torchbench_inference.csv"
+  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py --performance --only llama --backend eager --inference --devices mps --output "$TEST_REPORTS_DIR/torchbench_inference.csv"
+  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py --performance --only BERT_pytorch --backend eager --inference --devices mps --output "$TEST_REPORTS_DIR/torchbench_inference.csv"
+  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py --performance --only dcgan --backend eager --inference --devices mps --output "$TEST_REPORTS_DIR/torchbench_inference.csv"
+  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py --performance --only hf_GPT2 --backend eager --inference --devices mps --output "$TEST_REPORTS_DIR/torchbench_inference.csv"
+  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py --performance --only yolov3 --backend eager --inference --devices mps --output "$TEST_REPORTS_DIR/torchbench_inference.csv"
+  PYTHONPATH="$(pwd)"/torchbench python benchmarks/dynamo/torchbench.py --performance --only resnet152 --backend eager --inference --devices mps --output "$TEST_REPORTS_DIR/torchbench_inference.csv"
 
   echo "Pytorch benchmark on mps device completed"
 }
@@ -279,6 +267,25 @@ test_timm_perf() {
 
 install_tlparse
 
+if [[ $TEST_CONFIG == *"test_mps"* ]]; then
+  if [[ $NUM_TEST_SHARDS -gt 1 ]]; then
+    test_python_shard "${SHARD_NUMBER}"
+    if [[ "${SHARD_NUMBER}" == 1 ]]; then
+      test_libtorch
+      test_custom_script_ops
+    elif [[ "${SHARD_NUMBER}" == 2 ]]; then
+      test_jit_hooks
+      test_custom_backend
+    fi
+  else
+    test_python_all
+    test_libtorch
+    test_custom_script_ops
+    test_jit_hooks
+    test_custom_backend
+  fi
+fi
+
 if [[ $TEST_CONFIG == *"perf_all"* ]]; then
   test_torchbench_perf
   test_hf_perf
@@ -291,19 +298,4 @@ elif [[ $TEST_CONFIG == *"perf_timm"* ]]; then
   test_timm_perf
 elif [[ $TEST_CONFIG == *"perf_smoketest"* ]]; then
   test_torchbench_smoketest
-elif [[ $NUM_TEST_SHARDS -gt 1 ]]; then
-  test_python_shard "${SHARD_NUMBER}"
-  if [[ "${SHARD_NUMBER}" == 1 ]]; then
-    test_libtorch
-    test_custom_script_ops
-  elif [[ "${SHARD_NUMBER}" == 2 ]]; then
-    test_jit_hooks
-    test_custom_backend
-  fi
-else
-  test_python_all
-  test_libtorch
-  test_custom_script_ops
-  test_jit_hooks
-  test_custom_backend
 fi

.ci/pytorch/multigpu-test.sh

@@ -8,62 +8,55 @@
 source "$(dirname "${BASH_SOURCE[0]}")/common.sh"
 
 echo "Testing pytorch"
-# When adding more tests, please use HUD to see which shard is shorter
-if [[ "${SHARD_NUMBER:-1}" == "1" ]]; then
-    # FSDP tests
-    for f in test/distributed/fsdp/*.py ; do time python test/run_test.py --verbose -i "${f#*/}" ; done
-fi
+time python test/run_test.py --include test_cuda_multigpu test_cuda_primary_ctx --verbose
 
-if [[ "${SHARD_NUMBER:-2}" == "2" ]]; then
-    time python test/run_test.py --include test_cuda_multigpu test_cuda_primary_ctx --verbose
+# Disabling tests to see if they solve timeout issues; see https://github.com/pytorch/pytorch/issues/70015
+# python tools/download_mnist.py --quiet -d test/cpp/api/mnist
+# OMP_NUM_THREADS=2 TORCH_CPP_TEST_MNIST_PATH="test/cpp/api/mnist" build/bin/test_api
+time python test/run_test.py --verbose -i distributed/test_c10d_common
+time python test/run_test.py --verbose -i distributed/test_c10d_gloo
+time python test/run_test.py --verbose -i distributed/test_c10d_nccl
+time python test/run_test.py --verbose -i distributed/test_c10d_spawn_gloo
+time python test/run_test.py --verbose -i distributed/test_c10d_spawn_nccl
+time python test/run_test.py --verbose -i distributed/test_compute_comm_reordering
+time python test/run_test.py --verbose -i distributed/test_store
+time python test/run_test.py --verbose -i distributed/test_symmetric_memory
+time python test/run_test.py --verbose -i distributed/test_pg_wrapper
+time python test/run_test.py --verbose -i distributed/rpc/cuda/test_tensorpipe_agent
+# FSDP tests
+for f in test/distributed/fsdp/*.py ; do time python test/run_test.py --verbose -i "${f#*/}" ; done
+# ShardedTensor tests
+time python test/run_test.py --verbose -i distributed/checkpoint/test_checkpoint
+time python test/run_test.py --verbose -i distributed/checkpoint/test_file_system_checkpoint
+time python test/run_test.py --verbose -i distributed/_shard/sharding_spec/test_sharding_spec
+time python test/run_test.py --verbose -i distributed/_shard/sharding_plan/test_sharding_plan
+time python test/run_test.py --verbose -i distributed/_shard/sharded_tensor/test_sharded_tensor
+time python test/run_test.py --verbose -i distributed/_shard/sharded_tensor/test_sharded_tensor_reshard
 
-    # Disabling tests to see if they solve timeout issues; see https://github.com/pytorch/pytorch/issues/70015
-    # python tools/download_mnist.py --quiet -d test/cpp/api/mnist
-    # OMP_NUM_THREADS=2 TORCH_CPP_TEST_MNIST_PATH="test/cpp/api/mnist" build/bin/test_api
-    time python test/run_test.py --verbose -i distributed/test_c10d_common
-    time python test/run_test.py --verbose -i distributed/test_c10d_gloo
-    time python test/run_test.py --verbose -i distributed/test_c10d_nccl
-    time python test/run_test.py --verbose -i distributed/test_c10d_spawn_gloo
-    time python test/run_test.py --verbose -i distributed/test_c10d_spawn_nccl
-    time python test/run_test.py --verbose -i distributed/test_compute_comm_reordering
-    time python test/run_test.py --verbose -i distributed/test_store
-    time python test/run_test.py --verbose -i distributed/test_symmetric_memory
-    time python test/run_test.py --verbose -i distributed/test_pg_wrapper
-    time python test/run_test.py --verbose -i distributed/rpc/cuda/test_tensorpipe_agent
+# functional collective tests
+time python test/run_test.py --verbose -i distributed/test_functional_api
 
-    # ShardedTensor tests
-    time python test/run_test.py --verbose -i distributed/checkpoint/test_checkpoint
-    time python test/run_test.py --verbose -i distributed/checkpoint/test_file_system_checkpoint
-    time python test/run_test.py --verbose -i distributed/_shard/sharding_spec/test_sharding_spec
-    time python test/run_test.py --verbose -i distributed/_shard/sharding_plan/test_sharding_plan
-    time python test/run_test.py --verbose -i distributed/_shard/sharded_tensor/test_sharded_tensor
-    time python test/run_test.py --verbose -i distributed/_shard/sharded_tensor/test_sharded_tensor_reshard
+# DTensor tests
+time python test/run_test.py --verbose -i distributed/_tensor/test_random_ops
+time python test/run_test.py --verbose -i distributed/_tensor/test_dtensor_compile
 
-    # functional collective tests
-    time python test/run_test.py --verbose -i distributed/test_functional_api
+# DeviceMesh test
+time python test/run_test.py --verbose -i distributed/test_device_mesh
 
-    # DTensor tests
-    time python test/run_test.py --verbose -i distributed/tensor/test_random_ops
-    time python test/run_test.py --verbose -i distributed/tensor/test_dtensor_compile
+# DTensor/TP tests
+time python test/run_test.py --verbose -i distributed/tensor/parallel/test_tp_examples
+time python test/run_test.py --verbose -i distributed/tensor/parallel/test_tp_random_state
 
-    # DeviceMesh test
-    time python test/run_test.py --verbose -i distributed/test_device_mesh
+# FSDP2 tests
+time python test/run_test.py --verbose -i distributed/_composable/fsdp/test_fully_shard_training -- -k test_2d_mlp_with_nd_mesh
 
-    # DTensor/TP tests
-    time python test/run_test.py --verbose -i distributed/tensor/parallel/test_tp_examples
-    time python test/run_test.py --verbose -i distributed/tensor/parallel/test_tp_random_state
+# ND composability tests
+time python test/run_test.py --verbose -i distributed/_composable/test_composability/test_2d_composability
+time python test/run_test.py --verbose -i distributed/_composable/test_composability/test_pp_composability
 
-    # FSDP2 tests
-    time python test/run_test.py --verbose -i distributed/_composable/fsdp/test_fully_shard_training -- -k test_2d_mlp_with_nd_mesh
-
-    # ND composability tests
-    time python test/run_test.py --verbose -i distributed/_composable/test_composability/test_2d_composability
-    time python test/run_test.py --verbose -i distributed/_composable/test_composability/test_pp_composability
-
-    # Other tests
-    time python test/run_test.py --verbose -i test_cuda_primary_ctx
-    time python test/run_test.py --verbose -i test_optim -- -k test_forloop_goes_right_direction_multigpu
-    time python test/run_test.py --verbose -i test_optim -- -k test_mixed_device_dtype
-    time python test/run_test.py --verbose -i test_foreach -- -k test_tensors_grouping
-fi
+# Other tests
+time python test/run_test.py --verbose -i test_cuda_primary_ctx
+time python test/run_test.py --verbose -i test_optim -- -k test_forloop_goes_right_direction_multigpu
+time python test/run_test.py --verbose -i test_optim -- -k test_mixed_device_dtype
+time python test/run_test.py --verbose -i test_foreach -- -k test_tensors_grouping
 assert_git_not_dirty

.ci/pytorch/python_doc_push_script.sh

@@ -7,7 +7,7 @@ source "$pt_checkout/.ci/pytorch/common_utils.sh"
 
 echo "python_doc_push_script.sh: Invoked with $*"
 
-set -ex -o pipefail
+set -ex
 
 # for statements like ${1:-${DOCS_INSTALL_PATH:-docs/}}
 # the order of operations goes:
@@ -63,7 +63,7 @@ build_docs () {
     echo "(tried to echo the WARNINGS above the ==== line)"
     echo =========================
   fi
-  set -ex -o pipefail
+  set -ex
   return $code
 }
 

.ci/pytorch/run_tests.sh

@@ -1,436 +0,0 @@
-#!/bin/bash
-# shellcheck disable=SC2086,SC2048,SC2068,SC2145,SC2034,SC2207,SC2143
-# TODO: Re-enable shellchecks above
-
-set -eux -o pipefail
-
-# Essentially runs pytorch/test/run_test.py, but keeps track of which tests to
-# skip in a centralized place.
-#
-# TODO Except for a few tests, this entire file is a giant TODO. Why are these
-# tests # failing?
-# TODO deal with Windows
-
-# This script expects to be in the pytorch root folder
-if [[ ! -d 'test' || ! -f 'test/run_test.py' ]]; then
-    echo "run_tests.sh expects to be run from the Pytorch root directory " \
-         "but I'm actually in $(pwd)"
-    exit 2
-fi
-
-# Allow master skip of all tests
-if [[ -n "${SKIP_ALL_TESTS:-}" ]]; then
-    exit 0
-fi
-
-# If given specific test params then just run those
-if [[ -n "${RUN_TEST_PARAMS:-}" ]]; then
-    echo "$(date) :: Calling user-command $(pwd)/test/run_test.py ${RUN_TEST_PARAMS[@]}"
-    python test/run_test.py ${RUN_TEST_PARAMS[@]}
-    exit 0
-fi
-
-# Function to retry functions that sometimes timeout or have flaky failures
-retry () {
-    $*  || (sleep 1 && $*) || (sleep 2 && $*) || (sleep 4 && $*) || (sleep 8 && $*)
-}
-
-# Parameters
-##############################################################################
-if [[ "$#" != 3 ]]; then
-  if [[ -z "${DESIRED_PYTHON:-}" || -z "${DESIRED_CUDA:-}" || -z "${PACKAGE_TYPE:-}" ]]; then
-    echo "USAGE: run_tests.sh  PACKAGE_TYPE  DESIRED_PYTHON  DESIRED_CUDA"
-    echo "The env variable PACKAGE_TYPE must be set to 'manywheel' or 'libtorch'"
-    echo "The env variable DESIRED_PYTHON must be set like '2.7mu' or '3.6m' etc"
-    echo "The env variable DESIRED_CUDA must be set like 'cpu' or 'cu80' etc"
-    exit 1
-  fi
-  package_type="$PACKAGE_TYPE"
-  py_ver="$DESIRED_PYTHON"
-  cuda_ver="$DESIRED_CUDA"
-else
-  package_type="$1"
-  py_ver="$2"
-  cuda_ver="$3"
-fi
-
-if [[ "$cuda_ver" == 'cpu-cxx11-abi' ]]; then
-    cuda_ver="cpu"
-fi
-
-# cu80, cu90, cu100, cpu
-if [[ ${#cuda_ver} -eq 4 ]]; then
-    cuda_ver_majmin="${cuda_ver:2:1}.${cuda_ver:3:1}"
-elif [[ ${#cuda_ver} -eq 5 ]]; then
-    cuda_ver_majmin="${cuda_ver:2:2}.${cuda_ver:4:1}"
-fi
-
-NUMPY_PACKAGE=""
-if [[ ${py_ver} == "3.10" ]]; then
-    PROTOBUF_PACKAGE="protobuf>=3.17.2"
-    NUMPY_PACKAGE="numpy>=1.21.2"
-else
-    PROTOBUF_PACKAGE="protobuf=3.14.0"
-fi
-
-# Environment initialization
-if [[ "$(uname)" == Darwin ]]; then
-    # Install the testing dependencies
-    retry conda install -yq future hypothesis ${NUMPY_PACKAGE} ${PROTOBUF_PACKAGE} pytest setuptools six typing_extensions pyyaml
-else
-    retry pip install -qr requirements.txt || true
-    retry pip install -q hypothesis protobuf pytest setuptools || true
-    numpy_ver=1.15
-    case "$(python --version 2>&1)" in
-      *2* | *3.5* | *3.6*)
-        numpy_ver=1.11
-        ;;
-    esac
-    retry pip install -q "numpy==${numpy_ver}" || true
-fi
-
-echo "Testing with:"
-pip freeze
-conda list || true
-
-##############################################################################
-# Smoke tests
-##############################################################################
-# TODO use check_binary.sh, which requires making sure it runs on Windows
-pushd /
-echo "Smoke testing imports"
-python -c 'import torch'
-
-# Test that MKL is there
-if [[ "$(uname)" == 'Darwin' && "$package_type" == *wheel ]]; then
-    echo 'Not checking for MKL on Darwin wheel packages'
-else
-    echo "Checking that MKL is available"
-    python -c 'import torch; exit(0 if torch.backends.mkl.is_available() else 1)'
-fi
-
-if [[ "$OSTYPE" == "msys" ]]; then
-    GPUS=$(wmic path win32_VideoController get name)
-    if [[ ! "$GPUS" == *NVIDIA* ]]; then
-        echo "Skip CUDA tests for machines without a Nvidia GPU card"
-        exit 0
-    fi
-fi
-
-# Test that the version number is consistent during building and testing
-if [[ "$PYTORCH_BUILD_NUMBER" -gt 1 ]]; then
-    expected_version="${PYTORCH_BUILD_VERSION}.post${PYTORCH_BUILD_NUMBER}"
-else
-    expected_version="${PYTORCH_BUILD_VERSION}"
-fi
-echo "Checking that we are testing the package that is just built"
-python -c "import torch; exit(0 if torch.__version__ == '$expected_version' else 1)"
-
-# Test that CUDA builds are setup correctly
-if [[ "$cuda_ver" != 'cpu' ]]; then
-    cuda_installed=1
-    nvidia-smi || cuda_installed=0
-    if [[ "$cuda_installed" == 0 ]]; then
-      echo "Skip CUDA tests for machines without a Nvidia GPU card"
-    else
-      # Test CUDA archs
-      echo "Checking that CUDA archs are setup correctly"
-      timeout 20 python -c 'import torch; torch.randn([3,5]).cuda()'
-
-      # These have to run after CUDA is initialized
-      echo "Checking that magma is available"
-      python -c 'import torch; torch.rand(1).cuda(); exit(0 if torch.cuda.has_magma else 1)'
-      echo "Checking that CuDNN is available"
-      python -c 'import torch; exit(0 if torch.backends.cudnn.is_available() else 1)'
-    fi
-fi
-
-# Check that OpenBlas is not linked to on MacOS
-if [[ "$(uname)" == 'Darwin' ]]; then
-    echo "Checking the OpenBLAS is not linked to"
-    all_dylibs=($(find "$(python -c "import site; print(site.getsitepackages()[0])")"/torch -name '*.dylib'))
-    for dylib in "${all_dylibs[@]}"; do
-        if [[ -n "$(otool -L $dylib | grep -i openblas)" ]]; then
-            echo "Found openblas as a dependency of $dylib"
-            echo "Full dependencies is: $(otool -L $dylib)"
-            exit 1
-        fi
-    done
-
-    echo "Checking that OpenMP is available"
-    python -c "import torch; exit(0 if torch.backends.openmp.is_available() else 1)"
-fi
-
-popd
-
-# TODO re-enable the other tests after the nightlies are moved to CI. This is
-# because the binaries keep breaking, often from additional tests, that aren't
-# real problems. Once these are on circleci and a smoke-binary-build is added
-# to PRs then this should stop happening and these can be re-enabled.
-echo "Not running unit tests. Hopefully these problems are caught by CI"
-exit 0
-
-
-##############################################################################
-# Running unit tests (except not right now)
-##############################################################################
-echo "$(date) :: Starting tests for $package_type package for python$py_ver and $cuda_ver"
-
-# We keep track of exact tests to skip, as otherwise we would be hardly running
-# any tests. But b/c of issues working with pytest/normal-python-test/ and b/c
-# of special snowflake tests in test/run_test.py we also take special care of
-# those
-tests_to_skip=()
-
-#
-# Entire file exclusions
-##############################################################################
-entire_file_exclusions=("-x")
-
-# cpp_extensions doesn't work with pytest, so we exclude it from the pytest run
-# here and then manually run it later. Note that this is only because this
-# entire_fil_exclusions flag is only passed to the pytest run
-entire_file_exclusions+=("cpp_extensions")
-
-# TODO temporary line to fix next days nightlies, but should be removed when
-# issue is fixed
-entire_file_exclusions+=('type_info')
-
-if [[ "$cuda_ver" == 'cpu' ]]; then
-    # test/test_cuda.py exits early if the installed torch is not built with
-    # CUDA, but the exit doesn't work when running with pytest, so pytest will
-    # still try to run all the CUDA tests and then fail
-    entire_file_exclusions+=("cuda")
-    entire_file_exclusions+=("nccl")
-fi
-
-if [[ "$(uname)" == 'Darwin' || "$OSTYPE" == "msys" ]]; then
-    # pytest on Mac doesn't like the exits in these files
-    entire_file_exclusions+=('c10d')
-    entire_file_exclusions+=('distributed')
-
-    # pytest doesn't mind the exit but fails the tests. On Mac we run this
-    # later without pytest
-    entire_file_exclusions+=('thd_distributed')
-fi
-
-
-#
-# Universal flaky tests
-##############################################################################
-
-# RendezvousEnvTest sometimes hangs forever
-# Otherwise it will fail on CUDA with
-#   Traceback (most recent call last):
-#     File "test_c10d.py", line 179, in test_common_errors
-#       next(gen)
-#   AssertionError: ValueError not raised
-tests_to_skip+=('RendezvousEnvTest and test_common_errors')
-
-# This hung forever once on conda_3.5_cu92
-tests_to_skip+=('TestTorch and test_sum_dim')
-
-# test_trace_warn isn't actually flaky, but it doesn't work with pytest so we
-# just skip it
-tests_to_skip+=('TestJit and test_trace_warn')
-#
-# Python specific flaky tests
-##############################################################################
-
-# test_dataloader.py:721: AssertionError
-# looks like a timeout, but interestingly only appears on python 3
-if [[ "$py_ver" == 3* ]]; then
-    tests_to_skip+=('TestDataLoader and test_proper_exit')
-fi
-
-#
-# CUDA flaky tests, all package types
-##############################################################################
-if [[ "$cuda_ver" != 'cpu' ]]; then
-
-    #
-    # DistributedDataParallelTest
-    # All of these seem to fail
-    tests_to_skip+=('DistributedDataParallelTest')
-
-    #
-    # RendezvousEnvTest
-    # Traceback (most recent call last):
-    #   File "test_c10d.py", line 201, in test_nominal
-    #     store0, rank0, size0 = next(gen0)
-    #   File "/opt/python/cp36-cp36m/lib/python3.6/site-packages/torch/distributed/rendezvous.py", line 131, in _env_rendezvous_handler
-    #     store = TCPStore(master_addr, master_port, start_daemon)
-    # RuntimeError: Address already in use
-    tests_to_skip+=('RendezvousEnvTest and test_nominal')
-
-    #
-    # TestCppExtension
-    #
-    # Traceback (most recent call last):
-    #   File "test_cpp_extensions.py", line 134, in test_jit_cudnn_extension
-    #     with_cuda=True)
-    #   File "/opt/python/cp35-cp35m/lib/python3.5/site-packages/torch/utils/cpp_extension.py", line 552, in load
-    #     with_cuda)
-    #   File "/opt/python/cp35-cp35m/lib/python3.5/site-packages/torch/utils/cpp_extension.py", line 729, in _jit_compile
-    #     return _import_module_from_library(name, build_directory)
-    #   File "/opt/python/cp35-cp35m/lib/python3.5/site-packages/torch/utils/cpp_extension.py", line 867, in _import_module_from_library
-    #     return imp.load_module(module_name, file, path, description)
-    #   File "/opt/python/cp35-cp35m/lib/python3.5/imp.py", line 243, in load_module
-    #     return load_dynamic(name, filename, file)
-    #   File "/opt/python/cp35-cp35m/lib/python3.5/imp.py", line 343, in load_dynamic
-    #     return _load(spec)
-    #   File "<frozen importlib._bootstrap>", line 693, in _load
-    #   File "<frozen importlib._bootstrap>", line 666, in _load_unlocked
-    #   File "<frozen importlib._bootstrap>", line 577, in module_from_spec
-    #   File "<frozen importlib._bootstrap_external>", line 938, in create_module
-    #   File "<frozen importlib._bootstrap>", line 222, in _call_with_frames_removed
-    # ImportError: libcudnn.so.7: cannot open shared object file: No such file or directory
-    tests_to_skip+=('TestCppExtension and test_jit_cudnn_extension')
-
-    #
-    # TestCuda
-    #
-
-    # 3.7_cu80
-    #  RuntimeError: CUDA error: out of memory
-    tests_to_skip+=('TestCuda and test_arithmetic_large_tensor')
-
-    # 3.7_cu80
-    # RuntimeError: cuda runtime error (2) : out of memory at /opt/conda/conda-bld/pytorch-nightly_1538097262541/work/aten/src/THC/THCTensorCopy.cu:205
-    tests_to_skip+=('TestCuda and test_autogpu')
-
-    #
-    # TestDistBackend
-    #
-
-    # Traceback (most recent call last):
-    #   File "test_thd_distributed.py", line 1046, in wrapper
-    #     self._join_and_reduce(fn)
-    #   File "test_thd_distributed.py", line 1108, in _join_and_reduce
-    #     self.assertEqual(p.exitcode, first_process.exitcode)
-    #   File "/pytorch/test/common.py", line 399, in assertEqual
-    #     super(TestCase, self).assertEqual(x, y, message)
-    # AssertionError: None != 77 :
-    tests_to_skip+=('TestDistBackend and test_all_gather_group')
-    tests_to_skip+=('TestDistBackend and test_all_reduce_group_max')
-    tests_to_skip+=('TestDistBackend and test_all_reduce_group_min')
-    tests_to_skip+=('TestDistBackend and test_all_reduce_group_sum')
-    tests_to_skip+=('TestDistBackend and test_all_reduce_group_product')
-    tests_to_skip+=('TestDistBackend and test_barrier_group')
-    tests_to_skip+=('TestDistBackend and test_broadcast_group')
-
-    # Traceback (most recent call last):
-    #   File "test_thd_distributed.py", line 1046, in wrapper
-    #     self._join_and_reduce(fn)
-    #   File "test_thd_distributed.py", line 1108, in _join_and_reduce
-    #     self.assertEqual(p.exitcode, first_process.exitcode)
-    #   File "/pytorch/test/common.py", line 397, in assertEqual
-    #     super(TestCase, self).assertLessEqual(abs(x - y), prec, message)
-    # AssertionError: 12 not less than or equal to 1e-05
-    tests_to_skip+=('TestDistBackend and test_barrier')
-
-    # Traceback (most recent call last):
-    #   File "test_distributed.py", line 1267, in wrapper
-    #     self._join_and_reduce(fn)
-    #   File "test_distributed.py", line 1350, in _join_and_reduce
-    #     self.assertEqual(p.exitcode, first_process.exitcode)
-    #   File "/pytorch/test/common.py", line 399, in assertEqual
-    #     super(TestCase, self).assertEqual(x, y, message)
-    # AssertionError: None != 1
-    tests_to_skip+=('TestDistBackend and test_broadcast')
-
-    # Memory leak very similar to all the conda ones below, but appears on manywheel
-    # 3.6m_cu80
-    # AssertionError: 1605632 not less than or equal to 1e-05 : __main__.TestEndToEndHybridFrontendModels.test_vae_cuda leaked 1605632 bytes CUDA memory on device 0
-    tests_to_skip+=('TestEndToEndHybridFrontendModels and test_vae_cuda')
-
-    # ________________________ TestNN.test_embedding_bag_cuda ________________________
-    #
-    # self = <test_nn.TestNN testMethod=test_embedding_bag_cuda>
-    # dtype = torch.float32
-    #
-    #     @unittest.skipIf(not TEST_CUDA, "CUDA unavailable")
-    #     @repeat_test_for_types(ALL_TENSORTYPES)
-    #     @skipIfRocm
-    #     def test_embedding_bag_cuda(self, dtype=torch.float):
-    #         self._test_EmbeddingBag(True, 'sum', False, dtype)
-    #         self._test_EmbeddingBag(True, 'mean', False, dtype)
-    #         self._test_EmbeddingBag(True, 'max', False, dtype)
-    #         if dtype != torch.half:
-    #             # torch.cuda.sparse.HalfTensor is not enabled.
-    #             self._test_EmbeddingBag(True, 'sum', True, dtype)
-    # >           self._test_EmbeddingBag(True, 'mean', True, dtype)
-    #
-    # test_nn.py:2144:
-    # _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
-    # test_nn.py:2062: in _test_EmbeddingBag
-    #     _test_vs_Embedding(N, D, B, L)
-    # test_nn.py:2059: in _test_vs_Embedding
-    #     self.assertEqual(es_weight_grad, e.weight.grad, needed_prec)
-    # common.py:373: in assertEqual
-    #     assertTensorsEqual(x, y)
-    # common.py:365: in assertTensorsEqual
-    #     self.assertLessEqual(max_err, prec, message)
-    # E   AssertionError: tensor(0.0000, device='cuda:0', dtype=torch.float32) not less than or equal to 2e-05 :
-    #  1 failed, 1202 passed, 19 skipped, 2 xfailed, 796 warnings in 1166.73 seconds =
-    # Traceback (most recent call last):
-    #   File "test/run_test.py", line 391, in <module>
-    #     main()
-    #   File "test/run_test.py", line 383, in main
-    #     raise RuntimeError(message)
-    tests_to_skip+=('TestNN and test_embedding_bag_cuda')
-fi
-
-##############################################################################
-# MacOS specific flaky tests
-##############################################################################
-
-if [[ "$(uname)" == 'Darwin' ]]; then
-    # TestCppExtensions by default uses a temp folder in /tmp. This doesn't
-    # work for this Mac machine cause there is only one machine and /tmp is
-    # shared. (All the linux builds are on docker so have their own /tmp).
-    tests_to_skip+=('TestCppExtension')
-fi
-
-# Turn the set of tests to skip into an invocation that pytest understands
-excluded_tests_logic=''
-for exclusion in "${tests_to_skip[@]}"; do
-    if [[ -z "$excluded_tests_logic" ]]; then
-        # Only true for i==0
-        excluded_tests_logic="not ($exclusion)"
-    else
-        excluded_tests_logic="$excluded_tests_logic and not ($exclusion)"
-    fi
-done
-
-
-##############################################################################
-# Run the tests
-##############################################################################
-echo
-echo "$(date) :: Calling 'python test/run_test.py -v -p pytest ${entire_file_exclusions[@]} -- --disable-pytest-warnings -k '$excluded_tests_logic'"
-
-python test/run_test.py -v -p pytest ${entire_file_exclusions[@]} -- --disable-pytest-warnings -k "'" "$excluded_tests_logic" "'"
-
-echo
-echo "$(date) :: Finished 'python test/run_test.py -v -p pytest ${entire_file_exclusions[@]} -- --disable-pytest-warnings -k '$excluded_tests_logic'"
-
-# cpp_extensions don't work with pytest, so we run them without pytest here,
-# except there's a failure on CUDA builds (documented above), and
-# cpp_extensions doesn't work on a shared mac machine (also documented above)
-if [[ "$cuda_ver" == 'cpu' && "$(uname)" != 'Darwin' ]]; then
-    echo
-    echo "$(date) :: Calling 'python test/run_test.py -v -i cpp_extensions'"
-    python test/run_test.py -v -i cpp_extensions
-    echo
-    echo "$(date) :: Finished 'python test/run_test.py -v -i cpp_extensions'"
-fi
-
-# thd_distributed can run on Mac but not in pytest
-if [[ "$(uname)" == 'Darwin' ]]; then
-    echo
-    echo "$(date) :: Calling 'python test/run_test.py -v -i thd_distributed'"
-    python test/run_test.py -v -i thd_distributed
-    echo
-    echo "$(date) :: Finished 'python test/run_test.py -v -i thd_distributed'"
-fi

.ci/pytorch/smoke_test/check_binary_symbols.py

@@ -1,130 +0,0 @@
-#!/usr/bin/env python3
-import concurrent.futures
-import distutils.sysconfig
-import functools
-import itertools
-import os
-import re
-from pathlib import Path
-from typing import Any
-
-
-# We also check that there are [not] cxx11 symbols in libtorch
-#
-# To check whether it is using cxx11 ABI, check non-existence of symbol:
-PRE_CXX11_SYMBOLS = (
-    "std::basic_string<",
-    "std::list",
-)
-# To check whether it is using pre-cxx11 ABI, check non-existence of symbol:
-CXX11_SYMBOLS = (
-    "std::__cxx11::basic_string",
-    "std::__cxx11::list",
-)
-# NOTE: Checking the above symbols in all namespaces doesn't work, because
-# devtoolset7 always produces some cxx11 symbols even if we build with old ABI,
-# and CuDNN always has pre-cxx11 symbols even if we build with new ABI using gcc 5.4.
-# Instead, we *only* check the above symbols in the following namespaces:
-LIBTORCH_NAMESPACE_LIST = (
-    "c10::",
-    "at::",
-    "caffe2::",
-    "torch::",
-)
-
-
-def _apply_libtorch_symbols(symbols):
-    return [
-        re.compile(f"{x}.*{y}")
-        for (x, y) in itertools.product(LIBTORCH_NAMESPACE_LIST, symbols)
-    ]
-
-
-LIBTORCH_CXX11_PATTERNS = _apply_libtorch_symbols(CXX11_SYMBOLS)
-
-LIBTORCH_PRE_CXX11_PATTERNS = _apply_libtorch_symbols(PRE_CXX11_SYMBOLS)
-
-
-@functools.lru_cache(100)
-def get_symbols(lib: str) -> list[tuple[str, str, str]]:
-    from subprocess import check_output
-
-    lines = check_output(f'nm "{lib}"|c++filt', shell=True)
-    return [x.split(" ", 2) for x in lines.decode("latin1").split("\n")[:-1]]
-
-
-def grep_symbols(lib: str, patterns: list[Any]) -> list[str]:
-    def _grep_symbols(
-        symbols: list[tuple[str, str, str]], patterns: list[Any]
-    ) -> list[str]:
-        rc = []
-        for _s_addr, _s_type, s_name in symbols:
-            for pattern in patterns:
-                if pattern.match(s_name):
-                    rc.append(s_name)
-                    continue
-        return rc
-
-    all_symbols = get_symbols(lib)
-    num_workers = 32
-    chunk_size = (len(all_symbols) + num_workers - 1) // num_workers
-
-    def _get_symbols_chunk(i):
-        return all_symbols[i * chunk_size : (i + 1) * chunk_size]
-
-    with concurrent.futures.ThreadPoolExecutor(max_workers=32) as executor:
-        tasks = [
-            executor.submit(_grep_symbols, _get_symbols_chunk(i), patterns)
-            for i in range(num_workers)
-        ]
-        return functools.reduce(list.__add__, (x.result() for x in tasks), [])
-
-
-def check_lib_symbols_for_abi_correctness(lib: str, pre_cxx11_abi: bool = True) -> None:
-    print(f"lib: {lib}")
-    cxx11_symbols = grep_symbols(lib, LIBTORCH_CXX11_PATTERNS)
-    pre_cxx11_symbols = grep_symbols(lib, LIBTORCH_PRE_CXX11_PATTERNS)
-    num_cxx11_symbols = len(cxx11_symbols)
-    num_pre_cxx11_symbols = len(pre_cxx11_symbols)
-    print(f"num_cxx11_symbols: {num_cxx11_symbols}")
-    print(f"num_pre_cxx11_symbols: {num_pre_cxx11_symbols}")
-    if pre_cxx11_abi:
-        if num_cxx11_symbols > 0:
-            raise RuntimeError(
-                f"Found cxx11 symbols, but there shouldn't be any, see: {cxx11_symbols[:100]}"
-            )
-        if num_pre_cxx11_symbols < 1000:
-            raise RuntimeError("Didn't find enough pre-cxx11 symbols.")
-        # Check for no recursive iterators, regression test for https://github.com/pytorch/pytorch/issues/133437
-        rec_iter_symbols = grep_symbols(
-            lib, [re.compile("std::filesystem::recursive_directory_iterator.*")]
-        )
-        if len(rec_iter_symbols) > 0:
-            raise RuntimeError(
-                f"recursive_directory_iterator in used pre-CXX11 binaries, see; {rec_iter_symbols}"
-            )
-    else:
-        if num_pre_cxx11_symbols > 0:
-            raise RuntimeError(
-                f"Found pre-cxx11 symbols, but there shouldn't be any, see: {pre_cxx11_symbols[:100]}"
-            )
-        if num_cxx11_symbols < 100:
-            raise RuntimeError("Didn't find enought cxx11 symbols")
-
-
-def main() -> None:
-    if "install_root" in os.environ:
-        install_root = Path(os.getenv("install_root"))  # noqa: SIM112
-    else:
-        if os.getenv("PACKAGE_TYPE") == "libtorch":
-            install_root = Path(os.getcwd())
-        else:
-            install_root = Path(distutils.sysconfig.get_python_lib()) / "torch"
-
-    libtorch_cpu_path = install_root / "lib" / "libtorch_cpu.so"
-    pre_cxx11_abi = "cxx11-abi" not in os.getenv("DESIRED_DEVTOOLSET", "")
-    check_lib_symbols_for_abi_correctness(libtorch_cpu_path, pre_cxx11_abi)
-
-
-if __name__ == "__main__":
-    main()

.ci/pytorch/smoke_test/max_autotune.py

@@ -1,205 +0,0 @@
-import argparse
-
-from torchvision import datasets, transforms
-
-import torch
-import torch.nn as nn
-import torch.nn.functional as F
-import torch.optim as optim
-from torch.optim.lr_scheduler import StepLR
-
-
-class Net(nn.Module):
-    def __init__(self):
-        super(Net, self).__init__()  # noqa: UP008
-        self.conv1 = nn.Conv2d(1, 32, 3, 1)
-        self.conv2 = nn.Conv2d(32, 64, 3, 1)
-        self.dropout1 = nn.Dropout(0.25)
-        self.dropout2 = nn.Dropout(0.5)
-        self.fc1 = nn.Linear(9216, 128)
-        self.fc2 = nn.Linear(128, 10)
-
-    def forward(self, x):
-        x = self.conv1(x)
-        x = F.relu(x)
-        x = self.conv2(x)
-        x = F.relu(x)
-        x = F.max_pool2d(x, 2)
-        x = self.dropout1(x)
-        x = torch.flatten(x, 1)
-        x = self.fc1(x)
-        x = F.relu(x)
-        x = self.dropout2(x)
-        x = self.fc2(x)
-        output = F.log_softmax(x, dim=1)
-        return output
-
-
-def train(args, model, device, train_loader, optimizer, epoch):
-    model.train()
-    for batch_idx, (data, target) in enumerate(train_loader):
-        data, target = data.to(device), target.to(device)
-        optimizer.zero_grad()
-        output = model(data)
-        loss = F.nll_loss(output, target)
-        loss.backward()
-        optimizer.step()
-        if batch_idx % args.log_interval == 0:
-            print(
-                f"Train Epoch: {epoch} [{batch_idx * len(data)}/{len(train_loader.dataset)} ({100. * batch_idx / len(train_loader):.0f}%)]\tLoss: {loss.item():.6f}"  # noqa: B950
-            )
-            if args.dry_run:
-                break
-
-
-def test(model, device, test_loader):
-    model.eval()
-    test_loss = 0
-    correct = 0
-    with torch.no_grad():
-        for data, target in test_loader:
-            data, target = data.to(device), target.to(device)
-            output = model(data)
-            test_loss += F.nll_loss(
-                output, target, reduction="sum"
-            ).item()  # sum up batch loss
-            pred = output.argmax(
-                dim=1, keepdim=True
-            )  # get the index of the max log-probability
-            correct += pred.eq(target.view_as(pred)).sum().item()
-
-    test_loss /= len(test_loader.dataset)
-
-    print(
-        f"\nTest set: Average loss: {test_loss:.4f}, Accuracy: {correct}/{len(test_loader.dataset)} ({100. * correct / len(test_loader.dataset):.0f}%)\n"  # noqa: B950
-    )
-
-
-def timed(fn):
-    start = torch.cuda.Event(enable_timing=True)
-    end = torch.cuda.Event(enable_timing=True)
-    start.record()
-    result = fn()
-    end.record()
-    torch.cuda.synchronize()
-    return result, start.elapsed_time(end) / 1000
-
-
-def main():
-    # Training settings
-    parser = argparse.ArgumentParser(description="PyTorch MNIST Example")
-    parser.add_argument(
-        "--batch-size",
-        type=int,
-        default=64,
-        metavar="N",
-        help="input batch size for training (default: 64)",
-    )
-    parser.add_argument(
-        "--test-batch-size",
-        type=int,
-        default=1000,
-        metavar="N",
-        help="input batch size for testing (default: 1000)",
-    )
-    parser.add_argument(
-        "--epochs",
-        type=int,
-        default=4,
-        metavar="N",
-        help="number of epochs to train (default: 14)",
-    )
-    parser.add_argument(
-        "--lr",
-        type=float,
-        default=1.0,
-        metavar="LR",
-        help="learning rate (default: 1.0)",
-    )
-    parser.add_argument(
-        "--gamma",
-        type=float,
-        default=0.7,
-        metavar="M",
-        help="Learning rate step gamma (default: 0.7)",
-    )
-    parser.add_argument(
-        "--no-cuda", action="store_true", default=False, help="disables CUDA training"
-    )
-    parser.add_argument(
-        "--no-mps",
-        action="store_true",
-        default=False,
-        help="disables macOS GPU training",
-    )
-    parser.add_argument(
-        "--dry-run",
-        action="store_true",
-        default=False,
-        help="quickly check a single pass",
-    )
-    parser.add_argument(
-        "--seed", type=int, default=1, metavar="S", help="random seed (default: 1)"
-    )
-    parser.add_argument(
-        "--log-interval",
-        type=int,
-        default=100,
-        metavar="N",
-        help="how many batches to wait before logging training status",
-    )
-    parser.add_argument(
-        "--save-model",
-        action="store_true",
-        default=False,
-        help="For Saving the current Model",
-    )
-    args = parser.parse_args()
-    use_cuda = not args.no_cuda and torch.cuda.is_available()
-    use_mps = not args.no_mps and torch.backends.mps.is_available()
-
-    torch.manual_seed(args.seed)
-    torch.backends.cuda.matmul.allow_tf32 = True
-
-    if use_cuda:
-        device = torch.device("cuda")
-    elif use_mps:
-        device = torch.device("mps")
-    else:
-        device = torch.device("cpu")
-
-    train_kwargs = {"batch_size": args.batch_size}
-    test_kwargs = {"batch_size": args.test_batch_size}
-    if use_cuda:
-        cuda_kwargs = {"num_workers": 1, "pin_memory": True, "shuffle": True}
-        train_kwargs.update(cuda_kwargs)
-        test_kwargs.update(cuda_kwargs)
-
-    transform = transforms.Compose(
-        [transforms.ToTensor(), transforms.Normalize((0.1307,), (0.3081,))]
-    )
-    dataset1 = datasets.MNIST("../data", train=True, download=True, transform=transform)
-    dataset2 = datasets.MNIST("../data", train=False, transform=transform)
-    train_loader = torch.utils.data.DataLoader(dataset1, **train_kwargs)
-    test_loader = torch.utils.data.DataLoader(dataset2, **test_kwargs)
-
-    model = Net().to(device)
-    opt_model = torch.compile(model, mode="max-autotune")
-    optimizer = optim.Adadelta(opt_model.parameters(), lr=args.lr)
-
-    scheduler = StepLR(optimizer, step_size=1, gamma=args.gamma)
-    for epoch in range(1, args.epochs + 1):
-        print(
-            f"Training Time: {timed(lambda: train(args, opt_model, device, train_loader, optimizer, epoch))[1]}"
-        )
-        print(
-            f"Evaluation Time: {timed(lambda: test(opt_model, device, test_loader))[1]}"
-        )
-        scheduler.step()
-
-    if args.save_model:
-        torch.save(opt_model.state_dict(), "mnist_cnn.pt")
-
-
-if __name__ == "__main__":
-    main()

.ci/pytorch/smoke_test/smoke_test.py

@@ -1,422 +0,0 @@
-import argparse
-import importlib
-import json
-import os
-import re
-import subprocess
-import sys
-from pathlib import Path
-from tempfile import NamedTemporaryFile
-
-import torch
-import torch._dynamo
-import torch.nn as nn
-import torch.nn.functional as F
-
-
-if "MATRIX_GPU_ARCH_VERSION" in os.environ:
-    gpu_arch_ver = os.getenv("MATRIX_GPU_ARCH_VERSION")
-else:
-    gpu_arch_ver = os.getenv("GPU_ARCH_VERSION")  # Use fallback if available
-gpu_arch_type = os.getenv("MATRIX_GPU_ARCH_TYPE")
-channel = os.getenv("MATRIX_CHANNEL")
-package_type = os.getenv("MATRIX_PACKAGE_TYPE")
-target_os = os.getenv("TARGET_OS", sys.platform)
-BASE_DIR = Path(__file__).parent.parent.parent
-
-is_cuda_system = gpu_arch_type == "cuda"
-NIGHTLY_ALLOWED_DELTA = 3
-
-MODULES = [
-    {
-        "name": "torchvision",
-        "repo": "https://github.com/pytorch/vision.git",
-        "smoke_test": "./vision/test/smoke_test.py",
-        "extension": "extension",
-        "repo_name": "vision",
-    },
-    {
-        "name": "torchaudio",
-        "repo": "https://github.com/pytorch/audio.git",
-        "smoke_test": "./audio/test/smoke_test/smoke_test.py --no-ffmpeg",
-        "extension": "_extension",
-        "repo_name": "audio",
-    },
-]
-
-
-class Net(nn.Module):
-    def __init__(self):
-        super().__init__()
-        self.conv1 = nn.Conv2d(1, 32, 3, 1)
-        self.conv2 = nn.Conv2d(32, 64, 3, 1)
-        self.fc1 = nn.Linear(9216, 1)
-
-    def forward(self, x):
-        x = self.conv1(x)
-        x = self.conv2(x)
-        x = F.max_pool2d(x, 2)
-        x = torch.flatten(x, 1)
-        output = self.fc1(x)
-        return output
-
-
-def load_json_from_basedir(filename: str):
-    try:
-        with open(BASE_DIR / filename) as fptr:
-            return json.load(fptr)
-    except FileNotFoundError as exc:
-        raise ImportError(f"File {filename} not found error: {exc.strerror}") from exc
-    except json.JSONDecodeError as exc:
-        raise ImportError(f"Invalid JSON {filename}") from exc
-
-
-def read_release_matrix():
-    return load_json_from_basedir("release_matrix.json")
-
-
-def test_numpy():
-    import numpy as np
-
-    x = np.arange(5)
-    torch.tensor(x)
-
-
-def check_version(package: str) -> None:
-    release_version = os.getenv("RELEASE_VERSION")
-    # if release_version is specified, use it to validate the packages
-    if release_version:
-        release_matrix = read_release_matrix()
-        stable_version = release_matrix["torch"]
-    else:
-        stable_version = os.getenv("MATRIX_STABLE_VERSION")
-
-    # only makes sense to check nightly package where dates are known
-    if channel == "nightly":
-        check_nightly_binaries_date(package)
-    elif stable_version is not None:
-        if not torch.__version__.startswith(stable_version):
-            raise RuntimeError(
-                f"Torch version mismatch, expected {stable_version} for channel {channel}. But its {torch.__version__}"
-            )
-
-        if release_version and package == "all":
-            for module in MODULES:
-                imported_module = importlib.import_module(module["name"])
-                module_version = imported_module.__version__
-                if not module_version.startswith(release_matrix[module["name"]]):
-                    raise RuntimeError(
-                        f"{module['name']} version mismatch, expected: \
-                            {release_matrix[module['name']]} for channel {channel}. But its {module_version}"
-                    )
-                else:
-                    print(
-                        f"{module['name']} version actual: {module_version} expected: \
-                        {release_matrix[module['name']]} for channel {channel}."
-                    )
-
-    else:
-        print(f"Skip version check for channel {channel} as stable version is None")
-
-
-def check_nightly_binaries_date(package: str) -> None:
-    from datetime import datetime
-
-    format_dt = "%Y%m%d"
-
-    date_t_str = re.findall("dev\\d+", torch.__version__)
-    date_t_delta = datetime.now() - datetime.strptime(date_t_str[0][3:], format_dt)
-    if date_t_delta.days >= NIGHTLY_ALLOWED_DELTA:
-        raise RuntimeError(
-            f"the binaries are from {date_t_str} and are more than {NIGHTLY_ALLOWED_DELTA} days old!"
-        )
-
-    if package == "all":
-        for module in MODULES:
-            imported_module = importlib.import_module(module["name"])
-            module_version = imported_module.__version__
-            date_m_str = re.findall("dev\\d+", module_version)
-            date_m_delta = datetime.now() - datetime.strptime(
-                date_m_str[0][3:], format_dt
-            )
-            print(f"Nightly date check for {module['name']} version {module_version}")
-            if date_m_delta.days > NIGHTLY_ALLOWED_DELTA:
-                raise RuntimeError(
-                    f"Expected {module['name']} to be less then {NIGHTLY_ALLOWED_DELTA} days. But its {date_m_delta}"
-                )
-
-
-def test_cuda_runtime_errors_captured() -> None:
-    cuda_exception_missed = True
-    try:
-        print("Testing test_cuda_runtime_errors_captured")
-        torch._assert_async(torch.tensor(0, device="cuda"))
-        torch._assert_async(torch.tensor(0 + 0j, device="cuda"))
-    except RuntimeError as e:
-        if re.search("CUDA", f"{e}"):
-            print(f"Caught CUDA exception with success: {e}")
-            cuda_exception_missed = False
-        else:
-            raise e
-    if cuda_exception_missed:
-        raise RuntimeError("Expected CUDA RuntimeError but have not received!")
-
-
-def test_cuda_gds_errors_captured() -> None:
-    major_version = int(torch.version.cuda.split(".")[0])
-    minor_version = int(torch.version.cuda.split(".")[1])
-
-    if major_version < 12 or (major_version == 12 and minor_version < 6):
-        print("CUDA version is not supported for GDS smoke test")
-        return
-
-    cuda_exception_missed = True
-    try:
-        print("Testing test_cuda_gds_errors_captured")
-        with NamedTemporaryFile() as f:
-            torch.cuda.gds.GdsFile(f.name, os.O_CREAT | os.O_RDWR)
-    except RuntimeError as e:
-        expected_error = "cuFileHandleRegister failed"
-        if re.search(expected_error, f"{e}"):
-            print(f"Caught CUDA exception with success: {e}")
-            cuda_exception_missed = False
-        else:
-            raise e
-    if cuda_exception_missed:
-        raise RuntimeError(
-            "Expected cuFileHandleRegister failed RuntimeError but have not received!"
-        )
-
-
-def smoke_test_cuda(
-    package: str, runtime_error_check: str, torch_compile_check: str
-) -> None:
-    if not torch.cuda.is_available() and is_cuda_system:
-        raise RuntimeError(f"Expected CUDA {gpu_arch_ver}. However CUDA is not loaded.")
-
-    if package == "all" and is_cuda_system:
-        for module in MODULES:
-            imported_module = importlib.import_module(module["name"])
-            # TBD for vision move extension module to private so it will
-            # be _extention.
-            version = "N/A"
-            if module["extension"] == "extension":
-                version = imported_module.extension._check_cuda_version()
-            else:
-                version = imported_module._extension._check_cuda_version()
-            print(f"{module['name']} CUDA: {version}")
-
-    # torch.compile is available on macos-arm64 and Linux for python 3.8-3.13
-    if (
-        torch_compile_check == "enabled"
-        and sys.version_info < (3, 14, 0)
-        and target_os in ["linux", "linux-aarch64", "macos-arm64", "darwin"]
-    ):
-        smoke_test_compile("cuda" if torch.cuda.is_available() else "cpu")
-
-    if torch.cuda.is_available():
-        if torch.version.cuda != gpu_arch_ver:
-            raise RuntimeError(
-                f"Wrong CUDA version. Loaded: {torch.version.cuda} Expected: {gpu_arch_ver}"
-            )
-        print(f"torch cuda: {torch.version.cuda}")
-        # todo add cudnn version validation
-        print(f"torch cudnn: {torch.backends.cudnn.version()}")
-        print(f"cuDNN enabled? {torch.backends.cudnn.enabled}")
-
-        torch.cuda.init()
-        print("CUDA initialized successfully")
-        print(f"Number of CUDA devices: {torch.cuda.device_count()}")
-        for i in range(torch.cuda.device_count()):
-            print(f"Device {i}: {torch.cuda.get_device_name(i)}")
-
-        # nccl is availbale only on Linux
-        if sys.platform in ["linux", "linux2"]:
-            print(f"torch nccl version: {torch.cuda.nccl.version()}")
-
-        if runtime_error_check == "enabled":
-            test_cuda_runtime_errors_captured()
-
-
-def smoke_test_conv2d() -> None:
-    import torch.nn as nn
-
-    print("Testing smoke_test_conv2d")
-    # With square kernels and equal stride
-    m = nn.Conv2d(16, 33, 3, stride=2)
-    # non-square kernels and unequal stride and with padding
-    m = nn.Conv2d(16, 33, (3, 5), stride=(2, 1), padding=(4, 2))
-    assert m is not None
-    # non-square kernels and unequal stride and with padding and dilation
-    basic_conv = nn.Conv2d(
-        16, 33, (3, 5), stride=(2, 1), padding=(4, 2), dilation=(3, 1)
-    )
-    input = torch.randn(20, 16, 50, 100)
-    output = basic_conv(input)
-
-    if is_cuda_system:
-        print("Testing smoke_test_conv2d with cuda")
-        conv = nn.Conv2d(3, 3, 3).cuda()
-        x = torch.randn(1, 3, 24, 24, device="cuda")
-        with torch.cuda.amp.autocast():
-            out = conv(x)
-        assert out is not None
-
-        supported_dtypes = [torch.float16, torch.float32, torch.float64]
-        for dtype in supported_dtypes:
-            print(f"Testing smoke_test_conv2d with cuda for {dtype}")
-            conv = basic_conv.to(dtype).cuda()
-            input = torch.randn(20, 16, 50, 100, device="cuda").type(dtype)
-            output = conv(input)
-            assert output is not None
-
-
-def test_linalg(device="cpu") -> None:
-    print(f"Testing smoke_test_linalg on {device}")
-    A = torch.randn(5, 3, device=device)
-    U, S, Vh = torch.linalg.svd(A, full_matrices=False)
-    assert (
-        U.shape == A.shape
-        and S.shape == torch.Size([3])
-        and Vh.shape == torch.Size([3, 3])
-    )
-    torch.dist(A, U @ torch.diag(S) @ Vh)
-
-    U, S, Vh = torch.linalg.svd(A)
-    assert (
-        U.shape == torch.Size([5, 5])
-        and S.shape == torch.Size([3])
-        and Vh.shape == torch.Size([3, 3])
-    )
-    torch.dist(A, U[:, :3] @ torch.diag(S) @ Vh)
-
-    A = torch.randn(7, 5, 3, device=device)
-    U, S, Vh = torch.linalg.svd(A, full_matrices=False)
-    torch.dist(A, U @ torch.diag_embed(S) @ Vh)
-
-    if device == "cuda":
-        supported_dtypes = [torch.float32, torch.float64]
-        for dtype in supported_dtypes:
-            print(f"Testing smoke_test_linalg with cuda for {dtype}")
-            A = torch.randn(20, 16, 50, 100, device=device, dtype=dtype)
-            torch.linalg.svd(A)
-
-
-def smoke_test_compile(device: str = "cpu") -> None:
-    supported_dtypes = [torch.float16, torch.float32, torch.float64]
-
-    def foo(x: torch.Tensor) -> torch.Tensor:
-        return torch.sin(x) + torch.cos(x)
-
-    for dtype in supported_dtypes:
-        print(f"Testing smoke_test_compile for {device} and {dtype}")
-        x = torch.rand(3, 3, device=device).type(dtype)
-        x_eager = foo(x)
-        x_pt2 = torch.compile(foo)(x)
-        torch.testing.assert_close(x_eager, x_pt2)
-
-    # Check that SIMD were detected for the architecture
-    if device == "cpu":
-        from torch._inductor.codecache import pick_vec_isa
-
-        isa = pick_vec_isa()
-        if not isa:
-            raise RuntimeError("Can't detect vectorized ISA for CPU")
-        print(f"Picked CPU ISA {type(isa).__name__} bit width {isa.bit_width()}")
-
-    # Reset torch dynamo since we are changing mode
-    torch._dynamo.reset()
-    dtype = torch.float32
-    torch.set_float32_matmul_precision("high")
-    print(f"Testing smoke_test_compile with mode 'max-autotune' for {dtype}")
-    x = torch.rand(64, 1, 28, 28, device=device).type(torch.float32)
-    model = Net().to(device=device)
-    x_pt2 = torch.compile(model, mode="max-autotune")(x)
-
-
-def smoke_test_modules():
-    cwd = os.getcwd()
-    for module in MODULES:
-        if module["repo"]:
-            if not os.path.exists(f"{cwd}/{module['repo_name']}"):
-                print(f"Path does not exist: {cwd}/{module['repo_name']}")
-                try:
-                    subprocess.check_output(
-                        f"git clone --depth 1 {module['repo']}",
-                        stderr=subprocess.STDOUT,
-                        shell=True,
-                    )
-                except subprocess.CalledProcessError as exc:
-                    raise RuntimeError(
-                        f"Cloning {module['repo']} FAIL: {exc.returncode} Output: {exc.output}"
-                    ) from exc
-            try:
-                smoke_test_command = f"python3 {module['smoke_test']}"
-                if target_os == "windows":
-                    smoke_test_command = f"python {module['smoke_test']}"
-                output = subprocess.check_output(
-                    smoke_test_command,
-                    stderr=subprocess.STDOUT,
-                    shell=True,
-                    universal_newlines=True,
-                )
-            except subprocess.CalledProcessError as exc:
-                raise RuntimeError(
-                    f"Module {module['name']} FAIL: {exc.returncode} Output: {exc.output}"
-                ) from exc
-            else:
-                print(f"Output: \n{output}\n")
-
-
-def parse_args():
-    parser = argparse.ArgumentParser()
-    parser.add_argument(
-        "--package",
-        help="Package to include in smoke testing",
-        type=str,
-        choices=["all", "torchonly"],
-        default="all",
-    )
-    parser.add_argument(
-        "--runtime-error-check",
-        help="No Runtime Error check",
-        type=str,
-        choices=["enabled", "disabled"],
-        default="enabled",
-    )
-    parser.add_argument(
-        "--torch-compile-check",
-        help="Check torch compile",
-        type=str,
-        choices=["enabled", "disabled"],
-        default="enabled",
-    )
-    return parser.parse_args()
-
-
-def main() -> None:
-    options = parse_args()
-    print(f"torch: {torch.__version__}")
-    print(torch.__config__.parallel_info())
-    # All PyTorch binary builds should be built with OpenMP
-    if not torch.backends.openmp.is_available():
-        raise RuntimeError("PyTorch must be built with OpenMP support")
-
-    check_version(options.package)
-    smoke_test_conv2d()
-    test_linalg()
-    test_numpy()
-    if is_cuda_system:
-        test_linalg("cuda")
-        test_cuda_gds_errors_captured()
-
-    if options.package == "all":
-        smoke_test_modules()
-
-    smoke_test_cuda(
-        options.package, options.runtime_error_check, options.torch_compile_check
-    )
-
-
-if __name__ == "__main__":
-    main()

.ci/pytorch/test.sh

@@ -4,7 +4,7 @@
 # (This is set by default in the Docker images we build, so you don't
 # need to set it yourself.
 
-set -ex -o pipefail
+set -ex
 
 # Suppress ANSI color escape sequences
 export TERM=vt100
@@ -12,9 +12,9 @@ export TERM=vt100
 # shellcheck source=./common.sh
 source "$(dirname "${BASH_SOURCE[0]}")/common.sh"
 
-# Do not change workspace permissions for ROCm and s390x CI jobs
+# Do not change workspace permissions for ROCm CI jobs
 # as it can leave workspace with bad permissions for cancelled jobs
-if [[ "$BUILD_ENVIRONMENT" != *rocm* && "$BUILD_ENVIRONMENT" != *s390x* && -d /var/lib/jenkins/workspace ]]; then
+if [[ "$BUILD_ENVIRONMENT" != *rocm* ]]; then
   # Workaround for dind-rootless userid mapping (https://github.com/pytorch/ci-infra/issues/96)
   WORKSPACE_ORIGINAL_OWNER_ID=$(stat -c '%u' "/var/lib/jenkins/workspace")
   cleanup_workspace() {
@@ -46,9 +46,6 @@ BUILD_BIN_DIR="$BUILD_DIR"/bin
 SHARD_NUMBER="${SHARD_NUMBER:=1}"
 NUM_TEST_SHARDS="${NUM_TEST_SHARDS:=1}"
 
-# enable debug asserts in serialization
-export TORCH_SERIALIZATION_DEBUG=1
-
 export VALGRIND=ON
 # export TORCH_INDUCTOR_INSTALL_GXX=ON
 if [[ "$BUILD_ENVIRONMENT" == *clang9* || "$BUILD_ENVIRONMENT" == *xpu* ]]; then
@@ -89,13 +86,6 @@ if [[ "$BUILD_ENVIRONMENT" == *clang9* || "$BUILD_ENVIRONMENT" == *xpu* ]]; then
   export VALGRIND=OFF
 fi
 
-
-if [[ "$BUILD_ENVIRONMENT" == *s390x* ]]; then
-  # There are additional warnings on s390x, maybe due to newer gcc.
-  # Skip this check for now
-  export VALGRIND=OFF
-fi
-
 if [[ "${PYTORCH_TEST_RERUN_DISABLED_TESTS}" == "1" ]] || [[ "${CONTINUE_THROUGH_ERROR}" == "1" ]]; then
   # When rerunning disable tests, do not generate core dumps as it could consume
   # the runner disk space when crashed tests are run multiple times. Running out
@@ -139,7 +129,7 @@ if [[ "$TEST_CONFIG" == 'default' ]]; then
 fi
 
 if [[ "$TEST_CONFIG" == 'distributed' ]] && [[ "$BUILD_ENVIRONMENT" == *rocm* ]]; then
-  export HIP_VISIBLE_DEVICES=0,1,2,3
+  export HIP_VISIBLE_DEVICES=0,1
 fi
 
 if [[ "$TEST_CONFIG" == 'slow' ]]; then
@@ -163,8 +153,6 @@ elif [[ "$BUILD_ENVIRONMENT" == *xpu* ]]; then
   export PYTORCH_TESTING_DEVICE_ONLY_FOR="xpu"
   # setting PYTHON_TEST_EXTRA_OPTION
   export PYTHON_TEST_EXTRA_OPTION="--xpu"
-  # Disable sccache for xpu test due to flaky issue https://github.com/pytorch/pytorch/issues/143585
-  sudo rm -rf /opt/cache
 fi
 
 if [[ "$TEST_CONFIG" == *crossref* ]]; then
@@ -177,9 +165,6 @@ if [[ "$BUILD_ENVIRONMENT" == *rocm* ]]; then
   # Print GPU info
   rocminfo
   rocminfo | grep -E 'Name:.*\sgfx|Marketing'
-
-  # for benchmarks/dynamo/check_accuracy.py, we need to put results in a rocm specific directory to avoid clashes with cuda
-  MAYBE_ROCM="rocm/"
 fi
 
 if [[ "$BUILD_ENVIRONMENT" == *xpu* ]]; then
@@ -328,7 +313,6 @@ test_dynamo_wrapped_shard() {
     --exclude-jit-executor \
     --exclude-distributed-tests \
     --exclude-torch-export-tests \
-    --exclude-aot-dispatch-tests \
     --shard "$1" "$NUM_TEST_SHARDS" \
     --verbose \
     --upload-artifacts-while-running
@@ -342,7 +326,7 @@ test_inductor_distributed() {
   python test/run_test.py -i inductor/test_aot_inductor.py -k test_non_default_cuda_device --verbose
   python test/run_test.py -i inductor/test_aot_inductor.py -k test_replicate_on_devices --verbose
   python test/run_test.py -i distributed/test_c10d_functional_native.py --verbose
-  python test/run_test.py -i distributed/tensor/test_dtensor_compile.py --verbose
+  python test/run_test.py -i distributed/_tensor/test_dtensor_compile.py --verbose
   python test/run_test.py -i distributed/tensor/parallel/test_micro_pipeline_tp.py --verbose
   python test/run_test.py -i distributed/_composable/test_replicate_with_compiler.py --verbose
   python test/run_test.py -i distributed/_composable/fsdp/test_fully_shard_comm.py --verbose
@@ -395,32 +379,15 @@ test_inductor_aoti() {
   CPP_TESTS_DIR="${BUILD_BIN_DIR}" LD_LIBRARY_PATH="${TORCH_LIB_DIR}" python test/run_test.py --cpp --verbose -i cpp/test_aoti_abi_check cpp/test_aoti_inference
 }
 
-test_inductor_cpp_wrapper_shard() {
-  if [[ -z "$NUM_TEST_SHARDS" ]]; then
-    echo "NUM_TEST_SHARDS must be defined to run a Python test shard"
-    exit 1
-  fi
-
+test_inductor_cpp_wrapper() {
   export TORCHINDUCTOR_CPP_WRAPPER=1
   TEST_REPORTS_DIR=$(pwd)/test/test-reports
   mkdir -p "$TEST_REPORTS_DIR"
 
-  if [[ "$1" -eq "2" ]]; then
-    # For now, manually put the opinfo tests in shard 2, and all other tests in
-    # shard 1.  Test specific things triggering past bugs, for now.
-    python test/run_test.py \
-      --include inductor/test_torchinductor_opinfo \
-      -k 'linalg or to_sparse' \
-      --verbose
-    exit
-  fi
+  # Run certain inductor unit tests with cpp wrapper. In the end state, we should be able to run all the inductor
+  # unit tests with cpp wrapper.
+  python test/run_test.py --include inductor/test_torchinductor.py --verbose
 
-  # Run certain inductor unit tests with cpp wrapper. In the end state, we
-  # should be able to run all the inductor unit tests with cpp_wrapper.
-  python test/run_test.py \
-    --include inductor/test_torchinductor inductor/test_max_autotune inductor/test_cpu_repro \
-    --verbose
-  python test/run_test.py --inductor --include test_torch -k 'take' --verbose
 
   # Run inductor benchmark tests with cpp wrapper.
   # Skip benchmark tests if it's in rerun-disabled-mode.
@@ -433,7 +400,7 @@ test_inductor_cpp_wrapper_shard() {
     --output "$TEST_REPORTS_DIR/inductor_cpp_wrapper_training.csv"
     python benchmarks/dynamo/check_accuracy.py \
       --actual "$TEST_REPORTS_DIR/inductor_cpp_wrapper_training.csv" \
-      --expected "benchmarks/dynamo/ci_expected_accuracy/${MAYBE_ROCM}inductor_timm_training.csv"
+      --expected "benchmarks/dynamo/ci_expected_accuracy/inductor_timm_training.csv"
 
     python benchmarks/dynamo/torchbench.py --device cuda --accuracy \
       --bfloat16 --inference --inductor --only hf_T5 --output "$TEST_REPORTS_DIR/inductor_cpp_wrapper_inference.csv"
@@ -443,7 +410,7 @@ test_inductor_cpp_wrapper_shard() {
       --bfloat16 --inference --inductor --only moco --output "$TEST_REPORTS_DIR/inductor_cpp_wrapper_inference.csv"
     python benchmarks/dynamo/check_accuracy.py \
       --actual "$TEST_REPORTS_DIR/inductor_cpp_wrapper_inference.csv" \
-      --expected "benchmarks/dynamo/ci_expected_accuracy/${MAYBE_ROCM}inductor_torchbench_inference.csv"
+      --expected "benchmarks/dynamo/ci_expected_accuracy/inductor_torchbench_inference.csv"
   fi
 }
 
@@ -518,8 +485,6 @@ test_perf_for_dashboard() {
     test_inductor_set_cpu_affinity
   elif [[ "${TEST_CONFIG}" == *cuda_a10g* ]]; then
     device=cuda_a10g
-  elif [[ "${TEST_CONFIG}" == *rocm* ]]; then
-    device=rocm
   fi
 
   for mode in "${modes[@]}"; do
@@ -552,7 +517,7 @@ test_perf_for_dashboard() {
             --dynamic-batch-only "$@" \
             --output "$TEST_REPORTS_DIR/${backend}_dynamic_${suite}_${dtype}_${mode}_${device}_${target}.csv"
       fi
-      if [[ "$DASHBOARD_TAG" == *cppwrapper-true* ]]; then
+      if [[ "$DASHBOARD_TAG" == *cppwrapper-true* ]] && [[ "$mode" == "inference" ]]; then
         TORCHINDUCTOR_CPP_WRAPPER=1 $TASKSET python "benchmarks/dynamo/$suite.py" \
             "${target_flag[@]}" --"$mode" --"$dtype" --backend "$backend" --disable-cudagraphs "$@" \
             --output "$TEST_REPORTS_DIR/${backend}_cpp_wrapper_${suite}_${dtype}_${mode}_${device}_${target}.csv"
@@ -636,16 +601,16 @@ test_single_dynamo_benchmark() {
       TEST_CONFIG=${TEST_CONFIG//_avx512/}
     fi
     python "benchmarks/dynamo/$suite.py" \
-      --ci --accuracy --timing --explain --print-compilation-time \
+      --ci --accuracy --timing --explain \
       "${DYNAMO_BENCHMARK_FLAGS[@]}" \
       "$@" "${partition_flags[@]}" \
       --output "$TEST_REPORTS_DIR/${name}_${suite}.csv"
     python benchmarks/dynamo/check_accuracy.py \
       --actual "$TEST_REPORTS_DIR/${name}_$suite.csv" \
-      --expected "benchmarks/dynamo/ci_expected_accuracy/${MAYBE_ROCM}${TEST_CONFIG}_${name}.csv"
+      --expected "benchmarks/dynamo/ci_expected_accuracy/${TEST_CONFIG}_${name}.csv"
     python benchmarks/dynamo/check_graph_breaks.py \
       --actual "$TEST_REPORTS_DIR/${name}_$suite.csv" \
-      --expected "benchmarks/dynamo/ci_expected_accuracy/${MAYBE_ROCM}${TEST_CONFIG}_${name}.csv"
+      --expected "benchmarks/dynamo/ci_expected_accuracy/${TEST_CONFIG}_${name}.csv"
   fi
 }
 
@@ -668,7 +633,7 @@ test_inductor_halide() {
 }
 
 test_inductor_triton_cpu() {
-  python test/run_test.py --include inductor/test_triton_cpu_backend.py inductor/test_torchinductor_strided_blocks.py --verbose
+  python test/run_test.py --include inductor/test_triton_cpu_backend.py --verbose
   assert_git_not_dirty
 }
 
@@ -732,7 +697,7 @@ test_inductor_torchbench_smoketest_perf() {
       --only $test --output "$TEST_REPORTS_DIR/inductor_warm_start_smoketest_$test.csv"
     python benchmarks/dynamo/check_accuracy.py \
       --actual "$TEST_REPORTS_DIR/inductor_warm_start_smoketest_$test.csv" \
-      --expected "benchmarks/dynamo/ci_expected_accuracy/${MAYBE_ROCM}inductor_huggingface_training.csv"
+      --expected "benchmarks/dynamo/ci_expected_accuracy/inductor_huggingface_training.csv"
   done
 }
 
@@ -928,20 +893,10 @@ test_libtorch_api() {
   else
     # Exclude IMethodTest that relies on torch::deploy, which will instead be ran in test_deploy
     OMP_NUM_THREADS=2 TORCH_CPP_TEST_MNIST_PATH="${MNIST_DIR}" python test/run_test.py --cpp --verbose -i cpp/test_api -k "not IMethodTest"
-
-    # On s390x, pytorch is built without llvm.
-    # Even if it would be built with llvm, llvm currently doesn't support used features on s390x and
-    # test fails with errors like:
-    # JIT session error: Unsupported target machine architecture in ELF object pytorch-jitted-objectbuffer
-    # unknown file: Failure
-    # C++ exception with description "valOrErr INTERNAL ASSERT FAILED at "/var/lib/jenkins/workspace/torch/csrc/jit/tensorexpr/llvm_jit.h":34, please report a bug to PyTorch. Unexpected failure in LLVM JIT: Failed to materialize symbols: { (main, { func }) }
-    if [[ "${BUILD_ENVIRONMENT}" != *s390x* ]]; then
-      python test/run_test.py --cpp --verbose -i cpp/test_tensorexpr
-    fi
+    python test/run_test.py --cpp --verbose -i cpp/test_tensorexpr
   fi
 
-  # quantization is not fully supported on s390x yet
-  if [[ "${BUILD_ENVIRONMENT}" != *android* && "${BUILD_ENVIRONMENT}" != *cuda* && "${BUILD_ENVIRONMENT}" != *asan* && "${BUILD_ENVIRONMENT}" != *s390x* ]]; then
+  if [[ "${BUILD_ENVIRONMENT}" != *android* && "${BUILD_ENVIRONMENT}" != *cuda* && "${BUILD_ENVIRONMENT}" != *asan* ]]; then
     # NB: This test is not under TORCH_BIN_DIR but under BUILD_BIN_DIR
     export CPP_TESTS_DIR="${BUILD_BIN_DIR}"
     python test/run_test.py --cpp --verbose -i cpp/static_runtime_test
@@ -1288,7 +1243,7 @@ EOF
 }
 
 test_bazel() {
-  set -e -o pipefail
+  set -e
 
   # bazel test needs sccache setup.
   # shellcheck source=./common-build.sh
@@ -1439,7 +1394,7 @@ test_executorch() {
 test_linux_aarch64() {
   python test/run_test.py --include test_modules test_mkldnn test_mkldnn_fusion test_openmp test_torch test_dynamic_shapes \
         test_transformers test_multiprocessing test_numpy_interop test_autograd test_binary_ufuncs test_complex test_spectral_ops \
-        test_foreach test_reductions test_unary_ufuncs test_tensor_creation_ops \
+        test_foreach test_reductions test_unary_ufuncs \
         --shard "$SHARD_NUMBER" "$NUM_TEST_SHARDS" --verbose
 
   # Dynamo tests
@@ -1457,7 +1412,7 @@ test_linux_aarch64() {
        inductor/test_pattern_matcher inductor/test_perf inductor/test_profiler inductor/test_select_algorithm inductor/test_smoke \
        inductor/test_split_cat_fx_passes inductor/test_standalone_compile inductor/test_torchinductor \
        inductor/test_torchinductor_codegen_dynamic_shapes inductor/test_torchinductor_dynamic_shapes inductor/test_memory \
-       inductor/test_triton_cpu_backend inductor/test_triton_extension_backend inductor/test_mkldnn_pattern_matcher inductor/test_cpu_cpp_wrapper \
+       inductor/test_triton_cpu_backend inductor/test_triton_extension_backend \
        --shard "$SHARD_NUMBER" "$NUM_TEST_SHARDS" --verbose
 }
 
@@ -1466,9 +1421,9 @@ if ! [[ "${BUILD_ENVIRONMENT}" == *libtorch* || "${BUILD_ENVIRONMENT}" == *-baze
   (cd test && python -c "import torch; print(torch.__config__.parallel_info())")
 fi
 if [[ "${TEST_CONFIG}" == *numpy_2* ]]; then
-  # Install numpy-2.0.2 and compatible scipy & numba versions
-  python -mpip install --pre numpy==2.0.2 scipy==1.13.1 numba==0.60.0
-  python test/run_test.py --include dynamo/test_functions.py dynamo/test_unspec.py test_binary_ufuncs.py test_fake_tensor.py test_linalg.py test_numpy_interop.py test_tensor_creation_ops.py test_torch.py torch_np/test_basic.py
+  # Install numpy-2.0.2 and test inductor tracing
+  python -mpip install --pre numpy==2.0.2
+  python test/run_test.py --include dynamo/test_unspec.py
 elif [[ "${BUILD_ENVIRONMENT}" == *aarch64* && "${TEST_CONFIG}" != *perf_cpu_aarch64* ]]; then
   test_linux_aarch64
 elif [[ "${TEST_CONFIG}" == *backward* ]]; then
@@ -1542,7 +1497,7 @@ elif [[ "${TEST_CONFIG}" == *inductor_cpp_wrapper* ]]; then
   install_torchaudio cuda
   install_torchvision
   checkout_install_torchbench hf_T5 llama moco
-  PYTHONPATH=$(pwd)/torchbench test_inductor_cpp_wrapper_shard "$SHARD_NUMBER"
+  PYTHONPATH=$(pwd)/torchbench test_inductor_cpp_wrapper
 elif [[ "${TEST_CONFIG}" == *inductor* ]]; then
   install_torchvision
   test_inductor_shard "${SHARD_NUMBER}"

.ci/pytorch/test_example_code/CMakeLists.txt

@@ -1,26 +0,0 @@
-cmake_minimum_required(VERSION 3.0 FATAL_ERROR)
-project(simple-torch-test)
-
-find_package(Torch REQUIRED)
-set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${TORCH_CXX_FLAGS}")
-
-
-add_executable(simple-torch-test simple-torch-test.cpp)
-target_include_directories(simple-torch-test PRIVATE  ${TORCH_INCLUDE_DIRS})
-target_link_libraries(simple-torch-test "${TORCH_LIBRARIES}")
-set_property(TARGET simple-torch-test PROPERTY CXX_STANDARD 17)
-
-find_package(CUDAToolkit 11.8)
-
-target_link_libraries(simple-torch-test CUDA::cudart CUDA::cufft CUDA::cusparse CUDA::cublas CUDA::cusolver)
-find_library(CUDNN_LIBRARY NAMES cudnn)
-target_link_libraries(simple-torch-test  ${CUDNN_LIBRARY} )
-if(MSVC)
-  file(GLOB TORCH_DLLS  "$ENV{CUDA_PATH}/bin/cudnn64_8.dll" "$ENV{NVTOOLSEXT_PATH}/bin/x64/*.dll")
-  message("dlls to copy "  ${TORCH_DLLS})
-  add_custom_command(TARGET simple-torch-test
-                     POST_BUILD
-                     COMMAND ${CMAKE_COMMAND} -E copy_if_different
-                     ${TORCH_DLLS}
-                     $<TARGET_FILE_DIR:simple-torch-test>)
-endif(MSVC)

.ci/pytorch/test_example_code/check-torch-cuda.cpp

@@ -1,15 +0,0 @@
-#include <torch/torch.h>
-
-int main(int argc, const char* argv[]) {
-    std::cout << "Checking that CUDA archs are setup correctly" << std::endl;
-    TORCH_CHECK(torch::rand({ 3, 5 }, torch::Device(torch::kCUDA)).defined(), "CUDA archs are not setup correctly");
-
-    // These have to run after CUDA is initialized
-
-    std::cout << "Checking that magma is available" << std::endl;
-    TORCH_CHECK(torch::hasMAGMA(), "MAGMA is not available");
-
-    std::cout << "Checking that CuDNN is available" << std::endl;
-    TORCH_CHECK(torch::cuda::cudnn_is_available(), "CuDNN is not available");
-    return 0;
-}

.ci/pytorch/test_example_code/check-torch-mkl.cpp

@@ -1,6 +0,0 @@
-#include <torch/torch.h>
-
-int main(int argc, const char* argv[]) {
-    TORCH_CHECK(torch::hasMKL(), "MKL is not available");
-    return 0;
-}

.ci/pytorch/test_example_code/check-torch-xnnpack.cpp

@@ -1,7 +0,0 @@
-#include <ATen/ATen.h>
-#include <torch/torch.h>
-
-int main(int argc, const char* argv[]) {
-    TORCH_CHECK(at::globalContext().isXNNPACKAvailable(), "XNNPACK is not available");
-    return 0;
-}

.ci/pytorch/test_example_code/cnn_smoke.py

@@ -1,38 +0,0 @@
-r"""
-It's used to check basic rnn features with cuda.
-For example, it would throw exception if some components are missing
-"""
-
-import torch
-import torch.nn as nn
-import torch.nn.functional as F
-import torch.optim as optim
-
-
-class SimpleCNN(nn.Module):
-    def __init__(self):
-        super().__init__()
-        self.conv = nn.Conv2d(1, 1, 3)
-        self.pool = nn.MaxPool2d(2, 2)
-
-    def forward(self, inputs):
-        output = self.pool(F.relu(self.conv(inputs)))
-        output = output.view(1)
-        return output
-
-
-# Mock one infer
-device = torch.device("cuda:0")
-net = SimpleCNN().to(device)
-net_inputs = torch.rand((1, 1, 5, 5), device=device)
-outputs = net(net_inputs)
-print(outputs)
-
-criterion = nn.MSELoss()
-optimizer = optim.SGD(net.parameters(), lr=0.001, momentum=0.1)
-
-# Mock one step training
-label = torch.full((1,), 1.0, dtype=torch.float, device=device)
-loss = criterion(outputs, label)
-loss.backward()
-optimizer.step()

.ci/pytorch/test_example_code/cnn_smoke_win_arm64.py

@@ -1,41 +0,0 @@
-r"""
-It's used to check basic rnn features with cpu-only.
-For example, it would throw exception if some components are missing
-"""
-
-import torch
-import torch.nn as nn
-import torch.nn.functional as F
-import torch.optim as optim
-
-
-class SimpleCNN(nn.Module):
-    def __init__(self):
-        super().__init__()
-        self.conv = nn.Conv2d(1, 1, 3)
-        self.pool = nn.MaxPool2d(2, 2)
-
-    def forward(self, inputs):
-        output = self.pool(F.relu(self.conv(inputs)))
-        output = output.view(1)
-        return output
-
-
-try:
-    # Mock one infer
-    net = SimpleCNN()
-    net_inputs = torch.rand((1, 1, 5, 5))
-    outputs = net(net_inputs)
-    print(outputs)
-
-    criterion = nn.MSELoss()
-    optimizer = optim.SGD(net.parameters(), lr=0.001, momentum=0.1)
-
-    # Mock one step training
-    label = torch.full((1,), 1.0, dtype=torch.float)
-    loss = criterion(outputs, label)
-    loss.backward()
-    optimizer.step()
-
-except Exception as e:
-    print(f"An error occurred: {e}")

.ci/pytorch/test_example_code/rnn_smoke.py

@@ -1,14 +0,0 @@
-r"""
-It's used to check basic rnn features with cuda.
-For example, it would throw exception if missing some components are missing
-"""
-
-import torch
-import torch.nn as nn
-
-
-device = torch.device("cuda:0")
-rnn = nn.RNN(10, 20, 2).to(device)
-inputs = torch.randn(5, 3, 10).to(device)
-h0 = torch.randn(2, 3, 20).to(device)
-output, hn = rnn(inputs, h0)

.ci/pytorch/test_example_code/rnn_smoke_win_arm64.py

@@ -1,13 +0,0 @@
-r"""
-It's used to check basic rnn features with cpu-only.
-For example, it would throw exception if missing some components are missing
-"""
-
-import torch
-import torch.nn as nn
-
-
-rnn = nn.RNN(10, 20, 2)
-inputs = torch.randn(5, 3, 10)
-h0 = torch.randn(2, 3, 20)
-output, hn = rnn(inputs, h0)

.ci/pytorch/test_example_code/simple-torch-test.cpp

@@ -1,6 +0,0 @@
-#include <torch/torch.h>
-
-int main(int argc, const char* argv[]) {
-    TORCH_WARN("Simple test passed!");
-    return 0;
-}

.ci/pytorch/win-build.sh

@@ -38,7 +38,7 @@ if [[ $PYLONG_API_CHECK == 0 ]]; then
   echo "PyLong_AsUnsignedLong -> THPUtils_unpackUInt32 / THPUtils_unpackUInt64"
   exit 1
 fi
-set -ex -o pipefail
+set -ex
 
 "$SCRIPT_HELPERS_DIR"/build_pytorch.bat
 

.ci/pytorch/win-test-helpers/build_pytorch.bat

@@ -26,8 +26,7 @@ if not errorlevel 0 goto fail
 
 if "%USE_XPU%"=="1" (
   :: Install xpu support packages
-  set CUDA_VERSION=xpu
-  call %SCRIPT_HELPERS_DIR%\..\windows\internal\xpu_install.bat
+  call %INSTALLER_DIR%\install_xpu.bat
   if errorlevel 1 exit /b 1
 )
 

.ci/pytorch/win-test-helpers/installation-helpers/install_xpu.bat

@@ -2,14 +2,6 @@
 REM Description: Install Intel Support Packages on Windows
 REM BKM reference: https://www.intel.com/content/www/us/en/developer/articles/tool/pytorch-prerequisites-for-intel-gpus.html
 
-if not "%CUDA_VERSION%" == "xpu" (
-    echo Skipping for non XPU builds
-    exit /b 0
-)
-
-set SRC_DIR=%NIGHTLIES_PYTORCH_ROOT%
-if not exist "%SRC_DIR%\temp_build" mkdir "%SRC_DIR%\temp_build"
-
 set XPU_INSTALL_MODE=%~1
 if "%XPU_INSTALL_MODE%"=="" goto xpu_bundle_install_start
 if "%XPU_INSTALL_MODE%"=="bundle" goto xpu_bundle_install_start
@@ -120,14 +112,3 @@ if errorlevel 1 exit /b 1
 del xpu_extra.exe
 
 :xpu_install_end
-
-if not "%XPU_ENABLE_KINETO%"=="1" goto install_end
-:: Install Level Zero SDK
-set XPU_EXTRA_LZ_URL=https://github.com/oneapi-src/level-zero/releases/download/v1.14.0/level-zero-sdk_1.14.0.zip
-curl -k -L %XPU_EXTRA_LZ_URL% --output "%SRC_DIR%\temp_build\level_zero_sdk.zip"
-echo "Installing level zero SDK..."
-7z x "%SRC_DIR%\temp_build\level_zero_sdk.zip" -o"%SRC_DIR%\temp_build\level_zero"
-set "INCLUDE=%SRC_DIR%\temp_build\level_zero\include;%INCLUDE%"
-del "%SRC_DIR%\temp_build\level_zero_sdk.zip"
-
-:install_end

.ci/pytorch/win-test.sh

@@ -1,5 +1,5 @@
 #!/bin/bash
-set -ex -o pipefail
+set -ex
 
 SCRIPT_PARENT_DIR=$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )
 # shellcheck source=./common.sh
@@ -18,9 +18,6 @@ export PYTORCH_FINAL_PACKAGE_DIR="${PYTORCH_FINAL_PACKAGE_DIR:-/c/w/build-result
 PYTORCH_FINAL_PACKAGE_DIR_WIN=$(cygpath -w "${PYTORCH_FINAL_PACKAGE_DIR}")
 export PYTORCH_FINAL_PACKAGE_DIR_WIN
 
-# enable debug asserts in serialization
-export TORCH_SERIALIZATION_DEBUG=1
-
 mkdir -p "$TMP_DIR"/build/torch
 
 export SCRIPT_HELPERS_DIR=$SCRIPT_PARENT_DIR/win-test-helpers
@@ -38,13 +35,13 @@ if [[ "$BUILD_ENVIRONMENT" == *cuda* ]]; then
 fi
 
 # TODO: Move both of them to Windows AMI
-python -m pip install pytest-rerunfailures==10.3 pytest-cpp==2.3.0 tensorboard==2.13.0 pytest-subtests==0.13.1
+python -m pip install pytest-rerunfailures==10.3 pytest-cpp==2.3.0 tensorboard==2.13.0
 
 # Install Z3 optional dependency for Windows builds.
 python -m pip install z3-solver==4.12.2.0
 
 # Install tlparse for test\dynamo\test_structured_trace.py UTs.
-python -m pip install tlparse==0.3.30
+python -m pip install tlparse==0.3.25
 
 # Install parameterized
 python -m pip install parameterized==0.8.1

.ci/pytorch/windows/arm64/bootstrap_apl.bat

@@ -1,31 +0,0 @@
-@echo off
-
-echo Dependency ARM Performance Libraries (APL) installation started.
-
-:: Pre-check for downloads and dependencies folders
-if not exist "%DOWNLOADS_DIR%" mkdir %DOWNLOADS_DIR%
-if not exist "%DEPENDENCIES_DIR%" mkdir %DEPENDENCIES_DIR%
-
-:: Set download URL for the ARM Performance Libraries (APL)
-set DOWNLOAD_URL="https://developer.arm.com/-/cdn-downloads/permalink/Arm-Performance-Libraries/Version_24.10/arm-performance-libraries_24.10_Windows.msi"
-set INSTALLER_FILE=%DOWNLOADS_DIR%\arm-performance-libraries.msi
-
-:: Download installer
-echo Downloading ARM Performance Libraries (APL)...
-curl -L -o "%INSTALLER_FILE%" %DOWNLOAD_URL%
-
-:: Install ARM Performance Libraries (APL)
-echo Installing ARM Performance Libraries (APL)...
-msiexec /i "%INSTALLER_FILE%" /qn /norestart ACCEPT_EULA=1 INSTALLFOLDER="%DEPENDENCIES_DIR%"
-
-:: Check if installation was successful
-if %errorlevel% neq 0 (
-    echo "Failed to install ARM Performance Libraries (APL) components. (exitcode = %errorlevel%)"
-    exit /b 1
-)
-
-:: Add to environment
-echo ARMPL_DIR=%DEPENDENCIES_DIR%\armpl_24.10\>> %GITHUB_ENV%
-echo %DEPENDENCIES_DIR%\armpl_24.10\bin\>> %GITHUB_PATH%
-
-echo Dependency ARM Performance Libraries (APL) installation finished.

.ci/pytorch/windows/arm64/bootstrap_buildtools.bat

@@ -1,49 +0,0 @@
-@echo off
-
-echo Dependency MSVC Build Tools with C++ with ARM64/ARM64EC components installation started.
-
-:: Pre-check for downloads and dependencies folders
-if not exist "%DOWNLOADS_DIR%" mkdir "%DOWNLOADS_DIR%"
-if not exist "%DEPENDENCIES_DIR%" mkdir "%DEPENDENCIES_DIR%"
-
-:: Set download URL for the Visual Studio Installer
-set DOWNLOAD_URL=https://aka.ms/vs/17/release/vs_BuildTools.exe
-set INSTALLER_FILE=%DOWNLOADS_DIR%\vs_BuildTools.exe
-
-:: Download installer
-echo Downloading Visual Studio Build Tools with C++ installer...
-curl -L -o "%INSTALLER_FILE%" %DOWNLOAD_URL%
-
-:: Install the Visual Studio Build Tools with C++ components
-echo Installing Visual Studio Build Tools with C++ components...
-echo Installing MSVC %MSVC_VERSION%
-if "%MSVC_VERSION%" == "latest" (
-    "%INSTALLER_FILE%" --norestart --nocache --quiet --wait --installPath "%DEPENDENCIES_DIR%\VSBuildTools" ^
-        --add Microsoft.VisualStudio.Component.Windows11SDK.22621 ^
-        --add Microsoft.VisualStudio.Component.VC.ASAN ^
-        --add Microsoft.VisualStudio.Component.VC.CMake.Project ^
-        --add Microsoft.VisualStudio.Component.VC.Tools.ARM64 ^
-        --add Microsoft.VisualStudio.Component.VC.Tools.x86.x64
-) else if "%MSVC_VERSION%" == "14.40" (
-    "%INSTALLER_FILE%" --norestart --nocache --quiet --wait --installPath "%DEPENDENCIES_DIR%\VSBuildTools" ^
-        --add Microsoft.VisualStudio.Component.Windows11SDK.22621 ^
-        --add Microsoft.VisualStudio.Component.VC.ASAN ^
-        --add Microsoft.VisualStudio.Component.VC.CMake.Project ^
-        --add Microsoft.VisualStudio.Component.VC.14.40.17.10.ARM64 ^
-        --add Microsoft.VisualStudio.Component.VC.14.40.17.10.x86.x64
-) else if "%MSVC_VERSION%" == "14.36" (
-    "%INSTALLER_FILE%" --norestart --nocache --quiet --wait --installPath "%DEPENDENCIES_DIR%\VSBuildTools" ^
-        --add Microsoft.VisualStudio.Component.Windows11SDK.22621 ^
-        --add Microsoft.VisualStudio.Component.VC.ASAN ^
-        --add Microsoft.VisualStudio.Component.VC.CMake.Project ^
-        --add Microsoft.VisualStudio.Component.VC.14.36.17.6.ARM64 ^
-        --add Microsoft.VisualStudio.Component.VC.14.36.17.6.x86.x64
-)
-
-:: Check if installation was successful
-if %errorlevel% neq 0 (
-    echo "Failed to install Visual Studio Build Tools with C++ components. (exitcode = %errorlevel%)"
-    exit /b 1
-)
-
-echo Dependency Visual Studio Build Tools with C++ installation finished.

.ci/pytorch/windows/arm64/bootstrap_git.bat

@@ -1,37 +0,0 @@
-:: we need to install newer version of Git manually as "-submodules" function is not supported in the default version of runner.
-
-@echo off
-
-echo Dependency Git installation started.
-
-:: Pre-check for downloads and dependencies folders
-if not exist "%DOWNLOADS_DIR%" mkdir %DOWNLOADS_DIR%
-if not exist "%DEPENDENCIES_DIR%" mkdir %DEPENDENCIES_DIR%
-
-:: Set download URL for the Git
-set DOWNLOAD_URL="https://github.com/git-for-windows/git/releases/download/v2.46.0.windows.1/Git-2.46.0-64-bit.exe"
-set INSTALLER_FILE=%DOWNLOADS_DIR%\Git-2.46.0-64-bit.exe
-
-:: Download installer
-echo Downloading Git...
-curl -L -o "%INSTALLER_FILE%" %DOWNLOAD_URL%
-
-:: Install Git
-echo Installing Git...
-"%INSTALLER_FILE%" /VERYSILENT /DIR="%DEPENDENCIES_DIR%\git"
-
-dir %DEPENDENCIES_DIR%\git
-
-:: Check if installation was successful
-if %errorlevel% neq 0 (
-    echo "Failed to install Git. (exitcode = %errorlevel%)"
-    exit /b 1
-)
-
-:: Enable long paths
-call "%DEPENDENCIES_DIR%\git\cmd\git.exe" config --system core.longpaths true
-
-:: Add to PATH
-echo %DEPENDENCIES_DIR%\git\cmd\;%DEPENDENCIES_DIR%\git\bin\>> %GITHUB_PATH%
-
-echo Dependency Git installation finished.

.ci/pytorch/windows/arm64/bootstrap_libuv.bat

@@ -1,33 +0,0 @@
-@echo off
-
-echo Dependency libuv installation started.
-
-:: Pre-check for downloads and dependencies folders
-if not exist "%DOWNLOADS_DIR%" mkdir %DOWNLOADS_DIR%
-if not exist "%DEPENDENCIES_DIR%" mkdir %DEPENDENCIES_DIR%
-
-:: activate visual studio
-call "%DEPENDENCIES_DIR%\VSBuildTools\VC\Auxiliary\Build\vcvarsall.bat" arm64
-where cl.exe
-
-cd %DEPENDENCIES_DIR%
-git clone https://github.com/libuv/libuv.git -b v1.39.0
-
-echo Configuring libuv...
-mkdir libuv\build
-cd libuv\build
-cmake .. -DBUILD_TESTING=OFF
-
-echo Building libuv...
-cmake --build . --config Release
-
-echo Installing libuv...
-cmake --install . --prefix ../install
-
-:: Check if installation was successful
-if %errorlevel% neq 0 (
-    echo "Failed to install libuv. (exitcode = %errorlevel%)"
-    exit /b 1
-)
-
-echo Dependency libuv installation finished.

.ci/pytorch/windows/arm64/bootstrap_openblas.bat

@@ -1,46 +0,0 @@
-@echo off
-
-echo Dependency OpenBLAS installation started.
-
-:: Pre-check for downloads and dependencies folders
-if not exist "%DOWNLOADS_DIR%" mkdir %DOWNLOADS_DIR%
-if not exist "%DEPENDENCIES_DIR%" mkdir %DEPENDENCIES_DIR%
-
-:: activate visual studio
-call "%DEPENDENCIES_DIR%\VSBuildTools\VC\Auxiliary\Build\vcvarsall.bat" arm64
-where cl.exe
-
-:: Clone OpenBLAS
-cd %DEPENDENCIES_DIR%
-git clone https://github.com/OpenMathLib/OpenBLAS.git -b v0.3.29
-
-echo Configuring OpenBLAS...
-mkdir OpenBLAS\build
-cd OpenBLAS\build
-cmake .. -G Ninja ^
-  -DBUILD_TESTING=0 ^
-  -DBUILD_BENCHMARKS=0 ^
-  -DC_LAPACK=1 ^
-  -DNOFORTRAN=1 ^
-  -DDYNAMIC_ARCH=0 ^
-  -DARCH=arm64 ^
-  -DBINARY=64 ^
-  -DTARGET=GENERIC ^
-  -DUSE_OPENMP=1 ^
-  -DCMAKE_SYSTEM_PROCESSOR=ARM64 ^
-  -DCMAKE_SYSTEM_NAME=Windows ^
-  -DCMAKE_BUILD_TYPE=Release
-
-echo Building OpenBLAS...
-cmake --build . --config Release
-
-echo Installing OpenBLAS...
-cmake --install . --prefix ../install
-
-:: Check if installation was successful
-if %errorlevel% neq 0 (
-    echo "Failed to install OpenBLAS. (exitcode = %errorlevel%)"
-    exit /b 1
-)
-
-echo Dependency OpenBLAS installation finished.

.ci/pytorch/windows/arm64/bootstrap_python.bat

@@ -1,41 +0,0 @@
-@echo off
-
-echo Dependency Python installation started.
-
-:: Pre-check for downloads and dependencies folders
-if not exist "%DOWNLOADS_DIR%" mkdir %DOWNLOADS_DIR%
-if not exist "%DEPENDENCIES_DIR%" mkdir %DEPENDENCIES_DIR%
-
-if "%PYTHON_VERSION%"=="Python312" (
-    echo Python version is set to Python312
-    set DOWNLOAD_URL="https://www.python.org/ftp/python/3.12.7/python-3.12.7-arm64.exe"
-) else if "%PYTHON_VERSION%"=="Python311" (
-    echo Python version is set to Python311
-    set DOWNLOAD_URL="https://www.python.org/ftp/python/3.11.9/python-3.11.9-arm64.exe"
-) else (
-    echo PYTHON_VERSION not defined, Python version is set to Python312
-    set DOWNLOAD_URL="https://www.python.org/ftp/python/3.12.7/python-3.12.7-arm64.exe"
-)
-
-set INSTALLER_FILE=%DOWNLOADS_DIR%\python-installer.exe
-
-:: Download installer
-echo Downloading Python...
-curl -L -o "%INSTALLER_FILE%" %DOWNLOAD_URL%
-
-:: Install Python
-echo Installing Python...
-"%INSTALLER_FILE%" /quiet Include_debug=1 TargetDir="%DEPENDENCIES_DIR%\Python"
-
-:: Check if installation was successful
-if %errorlevel% neq 0 (
-    echo "Failed to install Python. (exitcode = %errorlevel%)"
-    exit /b 1
-)
-
-:: Add to PATH
-echo %DEPENDENCIES_DIR%\Python\>> %GITHUB_PATH%
-echo %DEPENDENCIES_DIR%\Python\scripts\>> %GITHUB_PATH%
-echo %DEPENDENCIES_DIR%\Python\libs\>> %GITHUB_PATH%
-
-echo Dependency Python installation finished.

.ci/pytorch/windows/arm64/bootstrap_rust.bat

@@ -1,33 +0,0 @@
-@echo off
-
-echo Dependency Rust installation started.
-
-:: Pre-check for downloads and dependencies folders
-if not exist "%DOWNLOADS_DIR%" mkdir %DOWNLOADS_DIR%
-if not exist "%DEPENDENCIES_DIR%" mkdir %DEPENDENCIES_DIR%
-
-set DOWNLOAD_URL="https://static.rust-lang.org/rustup/dist/x86_64-pc-windows-msvc/rustup-init.exe"
-set INSTALLER_FILE=%DOWNLOADS_DIR%\rustup-init.exe
-set RUSTUP_HOME=%DEPENDENCIES_DIR%\rust
-set CARGO_HOME=%DEPENDENCIES_DIR%\cargo
-
-:: Download installer
-echo Downloading Rust...
-curl -L -o "%INSTALLER_FILE%" %DOWNLOAD_URL%
-
-:: Install APL
-echo Installing Rust...
-"%INSTALLER_FILE%" -q -y --default-host aarch64-pc-windows-msvc --default-toolchain stable --profile default
-
-:: Check if installation was successful
-if %errorlevel% neq 0 (
-    echo "Failed to install Rust. (exitcode = %errorlevel%)"
-    exit /b 1
-)
-
-:: Add to PATH
-echo %DEPENDENCIES_DIR%\cargo\bin\>> %GITHUB_PATH%
-echo RUSTUP_HOME=%DEPENDENCIES_DIR%\rust>> %GITHUB_ENV%
-echo CARGO_HOME=%DEPENDENCIES_DIR%\cargo>> %GITHUB_ENV%
-
-echo Dependency Rust installation finished.

.ci/pytorch/windows/arm64/bootstrap_sccache.bat

@@ -1,33 +0,0 @@
-@echo off
-
-echo Dependency sccache installation started.
-
-:: Pre-check for downloads and dependencies folders
-if not exist "%DOWNLOADS_DIR%" mkdir %DOWNLOADS_DIR%
-if not exist "%DEPENDENCIES_DIR%" mkdir %DEPENDENCIES_DIR%
-
-:: Set download URL for the sccache
-set DOWNLOAD_URL="https://github.com/mozilla/sccache/releases/download/v0.8.1/sccache-v0.8.1-x86_64-pc-windows-msvc.zip"
-set INSTALLER_FILE=%DOWNLOADS_DIR%\sccache.zip
-
-:: Download installer
-echo Downloading sccache.zip...
-curl -L -o "%INSTALLER_FILE%" %DOWNLOAD_URL%
-
-:: Install sccache
-echo Extracting sccache.zip...
-tar -xf "%INSTALLER_FILE%" -C %DEPENDENCIES_DIR%
-cd %DEPENDENCIES_DIR%
-ren sccache-v0.8.1-x86_64-pc-windows-msvc sccache
-cd ..
-
-:: Check if installation was successful
-if %errorlevel% neq 0 (
-    echo "Failed to install sccache. (exitcode = %errorlevel%)"
-    exit /b 1
-)
-
-:: Add to PATH
-echo %DEPENDENCIES_DIR%\sccache\>> %GITHUB_PATH%
-
-echo Dependency sccache installation finished.

.ci/pytorch/windows/arm64/bootstrap_tests.bat

@@ -1,22 +0,0 @@
-:: change to source directory
-cd %PYTORCH_ROOT%
-
-:: activate visual studio
-call "%DEPENDENCIES_DIR%\VSBuildTools\VC\Auxiliary\Build\vcvarsall.bat" arm64
-where cl.exe
-
-:: create virtual environment
-python -m venv .venv
-echo * > .venv\.gitignore
-call .\.venv\Scripts\activate
-where python
-
-:: install dependencies
-python -m pip install --upgrade pip
-pip install -r requirements.txt
-pip install pytest numpy
-
-:: find file name for pytorch wheel
-for /f "delims=" %%f in ('dir /b "%PYTORCH_FINAL_PACKAGE_DIR%" ^| findstr "torch-"') do set "TORCH_WHEEL_FILENAME=%PYTORCH_FINAL_PACKAGE_DIR%\%%f"
-
-pip install %TORCH_WHEEL_FILENAME%

.ci/pytorch/windows/arm64/build_libtorch.bat

@@ -1,101 +0,0 @@
-@echo on
-
-:: environment variables
-set CMAKE_BUILD_TYPE=%BUILD_TYPE%
-set CMAKE_C_COMPILER_LAUNCHER=sccache
-set CMAKE_CXX_COMPILER_LAUNCHER=sccache
-set libuv_ROOT=%DEPENDENCIES_DIR%\libuv\install
-set MSSdk=1
-if defined PYTORCH_BUILD_VERSION (
-  set PYTORCH_BUILD_VERSION=%PYTORCH_BUILD_VERSION%
-  set PYTORCH_BUILD_NUMBER=1
-)
-
-:: Set BLAS type
-if %ENABLE_APL% == 1 (
-    set BLAS=APL
-    set USE_LAPACK=1
-) else if %ENABLE_OPENBLAS% == 1 (
-    set BLAS=OpenBLAS
-    set OpenBLAS_HOME=%DEPENDENCIES_DIR%\OpenBLAS\install
-)
-
-:: activate visual studio
-call "%DEPENDENCIES_DIR%\VSBuildTools\VC\Auxiliary\Build\vcvarsall.bat" arm64
-where cl.exe
-
-:: change to source directory
-cd %PYTORCH_ROOT%
-
-:: copy libuv.dll
-copy %libuv_ROOT%\lib\Release\uv.dll torch\lib\uv.dll
-
-:: create virtual environment
-python -m venv .venv
-echo * > .venv\.gitignore
-call .\.venv\Scripts\activate
-where python
-
-:: python install dependencies
-python -m pip install --upgrade pip
-pip install -r requirements.txt
-:: DISTUTILS_USE_SDK should be set after psutil dependency
-set DISTUTILS_USE_SDK=1
-
-:: start sccache server and reset sccache stats
-sccache --start-server
-sccache --zero-stats
-sccache --show-stats
- 
-:: Prepare the environment
-mkdir libtorch
-mkdir libtorch\bin
-mkdir libtorch\cmake
-mkdir libtorch\include
-mkdir libtorch\lib
-mkdir libtorch\share
-mkdir libtorch\test
-
-:: Call LibTorch build script
-python ./tools/build_libtorch.py
-
-:: Check if there is an error
-IF ERRORLEVEL 1 exit /b 1
-IF NOT ERRORLEVEL 0 exit /b 1
- 
-:: Move the files to the correct location
-move /Y torch\bin\*.* libtorch\bin\
-move /Y torch\cmake\*.* libtorch\cmake\
-robocopy /move /e torch\include\ libtorch\include\
-move /Y torch\lib\*.* libtorch\lib\
-robocopy /move /e torch\share\ libtorch\share\
-move /Y torch\test\*.* libtorch\test\
-move /Y libtorch\bin\*.dll libtorch\lib\
-
-:: Set version
-echo %PYTORCH_BUILD_VERSION% > libtorch\build-version
-git rev-parse HEAD > libtorch\build-hash
-
-:: Set LIBTORCH_PREFIX
-IF "%DEBUG%" == "" (
-    set LIBTORCH_PREFIX=libtorch-win-arm64-shared-with-deps
-) ELSE (
-    set LIBTORCH_PREFIX=libtorch-win-arm64-shared-with-deps-debug
-)
-
-:: Create output
-C:\Windows\System32\tar.exe -cvaf %LIBTORCH_PREFIX%-%PYTORCH_BUILD_VERSION%.zip -C libtorch *
-
-:: Copy output to target directory
-if not exist ..\output mkdir ..\output
-copy /Y "%LIBTORCH_PREFIX%-%PYTORCH_BUILD_VERSION%.zip" "%PYTORCH_FINAL_PACKAGE_DIR%\"
-copy /Y "%LIBTORCH_PREFIX%-%PYTORCH_BUILD_VERSION%.zip" "%PYTORCH_FINAL_PACKAGE_DIR%\%LIBTORCH_PREFIX%-latest.zip"
-
-:: Cleanup raw data to save space
-rmdir /s /q libtorch
-
-:: Check if installation was successful
-if %errorlevel% neq 0 (
-    echo "Failed on build_libtorch. (exitcode = %errorlevel%)"
-    exit /b 1
-)

.ci/pytorch/windows/arm64/build_pytorch.bat

@@ -1,60 +0,0 @@
-@echo on
-
-:: environment variables
-set CMAKE_BUILD_TYPE=%BUILD_TYPE%
-set CMAKE_C_COMPILER_LAUNCHER=sccache
-set CMAKE_CXX_COMPILER_LAUNCHER=sccache
-set libuv_ROOT=%DEPENDENCIES_DIR%\libuv\install
-set MSSdk=1
-if defined PYTORCH_BUILD_VERSION (
-  set PYTORCH_BUILD_VERSION=%PYTORCH_BUILD_VERSION%
-  set PYTORCH_BUILD_NUMBER=1
-)
-
-:: Set BLAS type
-if %ENABLE_APL% == 1 (
-    set BLAS=APL
-    set USE_LAPACK=1
-) else if %ENABLE_OPENBLAS% == 1 (
-    set BLAS=OpenBLAS
-    set OpenBLAS_HOME=%DEPENDENCIES_DIR%\OpenBLAS\install
-)
-
-:: activate visual studio
-call "%DEPENDENCIES_DIR%\VSBuildTools\VC\Auxiliary\Build\vcvarsall.bat" arm64
-where cl.exe
-
-:: change to source directory
-cd %PYTORCH_ROOT%
-
-:: copy libuv.dll
-copy %libuv_ROOT%\lib\Release\uv.dll torch\lib\uv.dll
-
-:: create virtual environment
-python -m venv .venv
-echo * > .venv\.gitignore
-call .\.venv\Scripts\activate
-where python
-
-:: python install dependencies
-python -m pip install --upgrade pip
-pip install -r requirements.txt
-:: DISTUTILS_USE_SDK should be set after psutil dependency
-set DISTUTILS_USE_SDK=1
-
-:: start sccache server and reset sccache stats
-sccache --start-server
-sccache --zero-stats
-sccache --show-stats
-
-:: Call PyTorch build script
-python setup.py bdist_wheel -d "%PYTORCH_FINAL_PACKAGE_DIR%"
-
-:: show sccache stats
-sccache --show-stats
-
-:: Check if installation was successful
-if %errorlevel% neq 0 (
-    echo "Failed on build_pytorch. (exitcode = %errorlevel%)"
-    exit /b 1
-)

.ci/pytorch/windows/arm64/smoke_test.bat

@@ -1,65 +0,0 @@
-@echo off
-setlocal
-
-set "ORIG_PATH=%PATH%"
-
-if "%PACKAGE_TYPE%" == "wheel" goto wheel
-if "%PACKAGE_TYPE%" == "libtorch" goto libtorch
-
-echo "unknown package type"
-exit /b 1
-
-:wheel
-echo "install wheel package"
-
-echo Running pip install...
-pip install -q --pre numpy protobuf
-echo Error level after pip install: %ERRORLEVEL%
-if errorlevel 1 exit /b 1
-
-for /F "delims=" %%i in ('where /R "%PYTORCH_FINAL_PACKAGE_DIR:/=\%" *.whl') do pip install "%%i"
-if errorlevel 1 exit /b 1
-
-goto smoke_test
-
-:smoke_test
-python -c "import torch"
-if ERRORLEVEL 1 exit /b 1
-
-echo Running python rnn_smoke.py...
-python %PYTORCH_ROOT%\.ci\pytorch\test_example_code\rnn_smoke_win_arm64.py
-if errorlevel 1 exit /b 1
-
-echo Checking that basic CNN works...
-python %PYTORCH_ROOT%\.ci\pytorch\test_example_code\cnn_smoke_win_arm64.py
-if errorlevel 1 exit /b 1
-
-goto end
-
-:libtorch
-echo "install and test libtorch"
-
-for /F "delims=" %%i in ('where /R "%PYTORCH_FINAL_PACKAGE_DIR:/=\%" *-latest.zip') do tar -xf "%%i" -C tmp
-if ERRORLEVEL 1 exit /b 1
-
-pushd tmp\libtorch
-
-set VC_VERSION_LOWER=14
-set VC_VERSION_UPPER=36
-
-call "%DEPENDENCIES_DIR%\VSBuildTools\VC\Auxiliary\Build\vcvarsall.bat" arm64
-
-set install_root=%CD%
-set INCLUDE=%INCLUDE%;%install_root%\include;%install_root%\include\torch\csrc\api\include
-set LIB=%LIB%;%install_root%\lib
-set PATH=%PATH%;%install_root%\lib
-
-cl %PYTORCH_ROOT%\.ci\pytorch\test_example_code\simple-torch-test.cpp c10.lib torch_cpu.lib /EHsc /std:c++17
-if ERRORLEVEL 1 exit /b 1
-
-.\simple-torch-test.exe
-if ERRORLEVEL 1 exit /b 1
-
-:end
-set "PATH=%ORIG_PATH%"
-popd

.ci/pytorch/windows/build_pytorch.bat

@@ -1,133 +0,0 @@
-@echo off
-
-:: This script parses args, installs required libraries (miniconda, MKL,
-:: Magma), and then delegates to cpu.bat, cuda80.bat, etc.
-
-if not "%CUDA_VERSION%" == "" if not "%PYTORCH_BUILD_VERSION%" == "" if not "%PYTORCH_BUILD_NUMBER%" == "" goto env_end
-if "%~1"=="" goto arg_error
-if "%~2"=="" goto arg_error
-if "%~3"=="" goto arg_error
-if not "%~4"=="" goto arg_error
-goto arg_end
-
-:arg_error
-
-echo Illegal number of parameters. Pass cuda version, pytorch version, build number
-echo CUDA version should be Mm with no dot, e.g. '80'
-echo DESIRED_PYTHON should be M.m, e.g. '2.7'
-exit /b 1
-
-:arg_end
-
-set CUDA_VERSION=%~1
-set PYTORCH_BUILD_VERSION=%~2
-set PYTORCH_BUILD_NUMBER=%~3
-
-:env_end
-
-set CUDA_PREFIX=cuda%CUDA_VERSION%
-if "%CUDA_VERSION%" == "cpu" set CUDA_PREFIX=cpu
-if "%CUDA_VERSION%" == "xpu" set CUDA_PREFIX=xpu
-
-if "%DESIRED_PYTHON%" == "" set DESIRED_PYTHON=3.5;3.6;3.7
-set DESIRED_PYTHON_PREFIX=%DESIRED_PYTHON:.=%
-set DESIRED_PYTHON_PREFIX=py%DESIRED_PYTHON_PREFIX:;=;py%
-
-set SRC_DIR=%~dp0
-pushd %SRC_DIR%
-
-:: Install Miniconda3
-set "CONDA_HOME=%CD%\conda"
-set "tmp_conda=%CONDA_HOME%"
-set "miniconda_exe=%CD%\miniconda.exe"
-rmdir /s /q conda
-del miniconda.exe
-curl --retry 3 -k https://repo.anaconda.com/miniconda/Miniconda3-py311_23.9.0-0-Windows-x86_64.exe -o "%miniconda_exe%"
-start /wait "" "%miniconda_exe%" /S /InstallationType=JustMe /RegisterPython=0 /AddToPath=0 /D=%tmp_conda%
-if ERRORLEVEL 1 exit /b 1
-set "ORIG_PATH=%PATH%"
-set "PATH=%CONDA_HOME%;%CONDA_HOME%\scripts;%CONDA_HOME%\Library\bin;%PATH%"
-
-:: create a new conda environment and install packages
-:try
-SET /A tries=3
-:loop
-IF %tries% LEQ 0 GOTO :exception
-call condaenv.bat
-IF %ERRORLEVEL% EQU 0 GOTO :done
-SET /A "tries=%tries%-1"
-:exception
-echo "Failed to create conda env"
-exit /B 1
-:done
-
-:: Download MAGMA Files on CUDA builds
-set MAGMA_VERSION=2.5.4
-
-if "%DEBUG%" == "1" (
-    set BUILD_TYPE=debug
-) else (
-    set BUILD_TYPE=release
-)
-
-if not "%CUDA_VERSION%" == "cpu" if not "%CUDA_VERSION%" == "xpu" (
-    rmdir /s /q magma_%CUDA_PREFIX%_%BUILD_TYPE%
-    del magma_%CUDA_PREFIX%_%BUILD_TYPE%.7z
-    curl -k https://s3.amazonaws.com/ossci-windows/magma_%MAGMA_VERSION%_%CUDA_PREFIX%_%BUILD_TYPE%.7z -o magma_%CUDA_PREFIX%_%BUILD_TYPE%.7z
-    7z x -aoa magma_%CUDA_PREFIX%_%BUILD_TYPE%.7z -omagma_%CUDA_PREFIX%_%BUILD_TYPE%
-)
-
-:: Install sccache
-if "%USE_SCCACHE%" == "1" (
-    mkdir %CD%\tmp_bin
-    curl -k https://s3.amazonaws.com/ossci-windows/sccache.exe --output %CD%\tmp_bin\sccache.exe
-    curl -k https://s3.amazonaws.com/ossci-windows/sccache-cl.exe --output %CD%\tmp_bin\sccache-cl.exe
-    if not "%CUDA_VERSION%" == "" (
-        set ADDITIONAL_PATH=%CD%\tmp_bin
-        set SCCACHE_IDLE_TIMEOUT=1500
-
-        :: randomtemp is used to resolve the intermittent build error related to CUDA.
-        :: code: https://github.com/peterjc123/randomtemp-rust
-        :: issue: https://github.com/pytorch/pytorch/issues/25393
-        ::
-        :: CMake requires a single command as CUDA_NVCC_EXECUTABLE, so we push the wrappers
-        :: randomtemp.exe and sccache.exe into a batch file which CMake invokes.
-        curl -kL https://github.com/peterjc123/randomtemp-rust/releases/download/v0.4/randomtemp.exe --output %SRC_DIR%\tmp_bin\randomtemp.exe
-        echo @"%SRC_DIR%\tmp_bin\randomtemp.exe" "%SRC_DIR%\tmp_bin\sccache.exe" "%CUDA_PATH%\bin\nvcc.exe" %%* > "%SRC_DIR%/tmp_bin/nvcc.bat"
-        cat %SRC_DIR%/tmp_bin/nvcc.bat
-        set CUDA_NVCC_EXECUTABLE=%SRC_DIR%/tmp_bin/nvcc.bat
-        :: CMake doesn't accept back-slashes in the path
-        for /F "usebackq delims=" %%n in (`cygpath -m "%CUDA_PATH%\bin\nvcc.exe"`) do set CMAKE_CUDA_COMPILER=%%n
-        set CMAKE_CUDA_COMPILER_LAUNCHER=%SRC_DIR%\tmp_bin\randomtemp.exe;%SRC_DIR%\tmp_bin\sccache.exe
-    )
-)
-
-set PYTORCH_BINARY_BUILD=1
-set TH_BINARY_BUILD=1
-set INSTALL_TEST=0
-
-for %%v in (%DESIRED_PYTHON_PREFIX%) do (
-    :: Activate Python Environment
-    set PYTHON_PREFIX=%%v
-    set "CONDA_LIB_PATH=%CONDA_HOME%\envs\%%v\Library\bin"
-    if not "%ADDITIONAL_PATH%" == "" (
-        set "PATH=%ADDITIONAL_PATH%;%CONDA_HOME%\envs\%%v;%CONDA_HOME%\envs\%%v\scripts;%CONDA_HOME%\envs\%%v\Library\bin;%ORIG_PATH%"
-    ) else (
-        set "PATH=%CONDA_HOME%\envs\%%v;%CONDA_HOME%\envs\%%v\scripts;%CONDA_HOME%\envs\%%v\Library\bin;%ORIG_PATH%"
-    )
-    pip install ninja
-    @setlocal
-    :: Set Flags
-    if not "%CUDA_VERSION%"=="cpu" if not "%CUDA_VERSION%" == "xpu" (
-        set MAGMA_HOME=%cd%\magma_%CUDA_PREFIX%_%BUILD_TYPE%
-    )
-    echo "Calling arch build script"
-    call %CUDA_PREFIX%.bat
-    if ERRORLEVEL 1 exit /b 1
-    @endlocal
-)
-
-set "PATH=%ORIG_PATH%"
-popd
-
-if ERRORLEVEL 1 exit /b 1

.ci/pytorch/windows/condaenv.bat

@@ -1,27 +0,0 @@
-IF "%DESIRED_PYTHON%"=="" (
-    echo DESIRED_PYTHON is NOT defined.
-    exit /b 1
-)
-
-:: Create a new conda environment
-setlocal EnableDelayedExpansion
-FOR %%v IN (%DESIRED_PYTHON%) DO (
-    set PYTHON_VERSION_STR=%%v
-    set PYTHON_VERSION_STR=!PYTHON_VERSION_STR:.=!
-    conda remove -n py!PYTHON_VERSION_STR! --all -y || rmdir %CONDA_HOME%\envs\py!PYTHON_VERSION_STR! /s
-    if "%%v" == "3.9" call conda create -n py!PYTHON_VERSION_STR! -y numpy=2.0.1 boto3 cmake ninja typing_extensions setuptools=72.1.0 python=%%v
-    if "%%v" == "3.10" call conda create -n py!PYTHON_VERSION_STR! -y -c=conda-forge numpy=2.0.1  boto3 cmake ninja typing_extensions setuptools=72.1.0 python=%%v
-    if "%%v" == "3.11" call conda create -n py!PYTHON_VERSION_STR! -y -c=conda-forge numpy=2.0.1  boto3 cmake ninja typing_extensions setuptools=72.1.0 python=%%v
-    if "%%v" == "3.12" call conda create -n py!PYTHON_VERSION_STR! -y -c=conda-forge numpy=2.0.1  boto3 cmake ninja typing_extensions setuptools=72.1.0 python=%%v
-    if "%%v" == "3.13" call conda create -n py!PYTHON_VERSION_STR! -y -c=conda-forge numpy=2.1.2  boto3 cmake ninja typing_extensions setuptools=72.1.0 python=%%v
-    if "%%v" == "3.13t" call conda create -n py!PYTHON_VERSION_STR! -y -c=conda-forge numpy=2.1.2 boto3 cmake ninja typing_extensions setuptools=72.1.0 python-freethreading python=3.13
-    call conda run -n py!PYTHON_VERSION_STR! pip install pyyaml
-    call conda run -n py!PYTHON_VERSION_STR! pip install mkl-include
-    call conda run -n py!PYTHON_VERSION_STR! pip install mkl-static
-)
-endlocal
-
-:: Install libuv
-conda install -y -q -c conda-forge libuv=1.39
-set libuv_ROOT=%CONDA_HOME%\Library
-echo libuv_ROOT=%libuv_ROOT%

.ci/pytorch/windows/cpu.bat

@@ -1,30 +0,0 @@
-@echo off
-
-set MODULE_NAME=pytorch
-
-IF NOT EXIST "setup.py" IF NOT EXIST "%MODULE_NAME%" (
-    call internal\clone.bat
-    cd %~dp0
-) ELSE (
-    call internal\clean.bat
-)
-
-IF ERRORLEVEL 1 goto :eof
-
-call internal\check_deps.bat
-IF ERRORLEVEL 1 goto :eof
-
-REM Check for optional components
-
-echo Disabling CUDA
-set USE_CUDA=0
-
-call internal\check_opts.bat
-IF ERRORLEVEL 1 goto :eof
-
-if exist "%NIGHTLIES_PYTORCH_ROOT%" cd %NIGHTLIES_PYTORCH_ROOT%\..
-call %~dp0\internal\copy_cpu.bat
-IF ERRORLEVEL 1 goto :eof
-
-call %~dp0\internal\setup.bat
-IF ERRORLEVEL 1 goto :eof

.ci/pytorch/windows/cuda118.bat

@@ -1,59 +0,0 @@
-@echo off
-
-set MODULE_NAME=pytorch
-
-IF NOT EXIST "setup.py" IF NOT EXIST "%MODULE_NAME%" (
-    call internal\clone.bat
-    cd %~dp0
-) ELSE (
-    call internal\clean.bat
-)
-IF ERRORLEVEL 1 goto :eof
-
-call internal\check_deps.bat
-IF ERRORLEVEL 1 goto :eof
-
-REM Check for optional components
-
-set USE_CUDA=
-set CMAKE_GENERATOR=Visual Studio 15 2017 Win64
-
-IF "%NVTOOLSEXT_PATH%"=="" (
-    IF EXIST "C:\Program Files\NVIDIA Corporation\NvToolsExt\lib\x64\nvToolsExt64_1.lib"  (
-        set NVTOOLSEXT_PATH=C:\Program Files\NVIDIA Corporation\NvToolsExt
-    ) ELSE (
-        echo NVTX ^(Visual Studio Extension ^for CUDA^) ^not installed, failing
-        exit /b 1
-    )
-)
-
-IF "%CUDA_PATH_V118%"=="" (
-    IF EXIST "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.8\bin\nvcc.exe" (
-        set "CUDA_PATH_V118=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.8"
-    ) ELSE (
-        echo CUDA 11.8 not found, failing
-        exit /b 1
-    )
-)
-
-IF "%BUILD_VISION%" == "" (
-    set TORCH_CUDA_ARCH_LIST=3.7+PTX;5.0;6.0;6.1;7.0;7.5;8.0;8.6;9.0
-    set TORCH_NVCC_FLAGS=-Xfatbin -compress-all
-) ELSE (
-    set NVCC_FLAGS=-D__CUDA_NO_HALF_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_35,code=sm_35 -gencode=arch=compute_50,code=sm_50 -gencode=arch=compute_60,code=sm_60 -gencode=arch=compute_70,code=sm_70 -gencode=arch=compute_75,code=sm_75 -gencode=arch=compute_80,code=compute_80 -gencode=arch=compute_86,code=compute_86 -gencode=arch=compute_90,code=compute_90
-)
-
-set "CUDA_PATH=%CUDA_PATH_V118%"
-set "PATH=%CUDA_PATH_V118%\bin;%PATH%"
-
-:optcheck
-
-call internal\check_opts.bat
-IF ERRORLEVEL 1 goto :eof
-
-if exist "%NIGHTLIES_PYTORCH_ROOT%" cd %NIGHTLIES_PYTORCH_ROOT%\..
-call  %~dp0\internal\copy.bat
-IF ERRORLEVEL 1 goto :eof
-
-call  %~dp0\internal\setup.bat
-IF ERRORLEVEL 1 goto :eof

.ci/pytorch/windows/cuda124.bat

@@ -1,59 +0,0 @@
-@echo off
-
-set MODULE_NAME=pytorch
-
-IF NOT EXIST "setup.py" IF NOT EXIST "%MODULE_NAME%" (
-    call internal\clone.bat
-    cd %~dp0
-) ELSE (
-    call internal\clean.bat
-)
-IF ERRORLEVEL 1 goto :eof
-
-call internal\check_deps.bat
-IF ERRORLEVEL 1 goto :eof
-
-REM Check for optional components
-
-set USE_CUDA=
-set CMAKE_GENERATOR=Visual Studio 15 2017 Win64
-
-IF "%NVTOOLSEXT_PATH%"=="" (
-    IF EXIST "C:\Program Files\NVIDIA Corporation\NvToolsExt\lib\x64\nvToolsExt64_1.lib"  (
-        set NVTOOLSEXT_PATH=C:\Program Files\NVIDIA Corporation\NvToolsExt
-    ) ELSE (
-        echo NVTX ^(Visual Studio Extension ^for CUDA^) ^not installed, failing
-        exit /b 1
-    )
-)
-
-IF "%CUDA_PATH_V124%"=="" (
-    IF EXIST "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4\bin\nvcc.exe" (
-        set "CUDA_PATH_V124=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4"
-    ) ELSE (
-        echo CUDA 12.4 not found, failing
-        exit /b 1
-    )
-)
-
-IF "%BUILD_VISION%" == "" (
-    set TORCH_CUDA_ARCH_LIST=5.0;6.0;6.1;7.0;7.5;8.0;8.6;9.0
-    set TORCH_NVCC_FLAGS=-Xfatbin -compress-all
-) ELSE (
-    set NVCC_FLAGS=-D__CUDA_NO_HALF_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_50,code=sm_50 -gencode=arch=compute_60,code=sm_60 -gencode=arch=compute_70,code=sm_70 -gencode=arch=compute_75,code=sm_75 -gencode=arch=compute_80,code=compute_80 -gencode=arch=compute_86,code=compute_86 -gencode=arch=compute_90,code=compute_90
-)
-
-set "CUDA_PATH=%CUDA_PATH_V124%"
-set "PATH=%CUDA_PATH_V124%\bin;%PATH%"
-
-:optcheck
-
-call internal\check_opts.bat
-IF ERRORLEVEL 1 goto :eof
-
-if exist "%NIGHTLIES_PYTORCH_ROOT%" cd %NIGHTLIES_PYTORCH_ROOT%\..
-call  %~dp0\internal\copy.bat
-IF ERRORLEVEL 1 goto :eof
-
-call  %~dp0\internal\setup.bat
-IF ERRORLEVEL 1 goto :eof

.ci/pytorch/windows/cuda126.bat

@@ -1,59 +0,0 @@
-@echo off
-
-set MODULE_NAME=pytorch
-
-IF NOT EXIST "setup.py" IF NOT EXIST "%MODULE_NAME%" (
-    call internal\clone.bat
-    cd %~dp0
-) ELSE (
-    call internal\clean.bat
-)
-IF ERRORLEVEL 1 goto :eof
-
-call internal\check_deps.bat
-IF ERRORLEVEL 1 goto :eof
-
-REM Check for optional components
-
-set USE_CUDA=
-set CMAKE_GENERATOR=Visual Studio 15 2017 Win64
-
-IF "%NVTOOLSEXT_PATH%"=="" (
-    IF EXIST "C:\Program Files\NVIDIA Corporation\NvToolsExt\lib\x64\nvToolsExt64_1.lib"  (
-        set NVTOOLSEXT_PATH=C:\Program Files\NVIDIA Corporation\NvToolsExt
-    ) ELSE (
-        echo NVTX ^(Visual Studio Extension ^for CUDA^) ^not installed, failing
-        exit /b 1
-    )
-)
-
-IF "%CUDA_PATH_V126%"=="" (
-    IF EXIST "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.6\bin\nvcc.exe" (
-        set "CUDA_PATH_V126=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.6"
-    ) ELSE (
-        echo CUDA 12.6 not found, failing
-        exit /b 1
-    )
-)
-
-IF "%BUILD_VISION%" == "" (
-    set TORCH_CUDA_ARCH_LIST=5.0;6.0;6.1;7.0;7.5;8.0;8.6;9.0
-    set TORCH_NVCC_FLAGS=-Xfatbin -compress-all
-) ELSE (
-    set NVCC_FLAGS=-D__CUDA_NO_HALF_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_50,code=sm_50 -gencode=arch=compute_60,code=sm_60 -gencode=arch=compute_70,code=sm_70 -gencode=arch=compute_75,code=sm_75 -gencode=arch=compute_80,code=compute_80 -gencode=arch=compute_86,code=compute_86 -gencode=arch=compute_90,code=compute_90
-)
-
-set "CUDA_PATH=%CUDA_PATH_V126%"
-set "PATH=%CUDA_PATH_V126%\bin;%PATH%"
-
-:optcheck
-
-call internal\check_opts.bat
-IF ERRORLEVEL 1 goto :eof
-
-if exist "%NIGHTLIES_PYTORCH_ROOT%" cd %NIGHTLIES_PYTORCH_ROOT%\..
-call  %~dp0\internal\copy.bat
-IF ERRORLEVEL 1 goto :eof
-
-call  %~dp0\internal\setup.bat
-IF ERRORLEVEL 1 goto :eof

.ci/pytorch/windows/cuda128.bat

@@ -1,59 +0,0 @@
-@echo off
-
-set MODULE_NAME=pytorch
-
-IF NOT EXIST "setup.py" IF NOT EXIST "%MODULE_NAME%" (
-    call internal\clone.bat
-    cd %~dp0
-) ELSE (
-    call internal\clean.bat
-)
-IF ERRORLEVEL 1 goto :eof
-
-call internal\check_deps.bat
-IF ERRORLEVEL 1 goto :eof
-
-REM Check for optional components
-
-set USE_CUDA=
-set CMAKE_GENERATOR=Visual Studio 15 2017 Win64
-
-IF "%NVTOOLSEXT_PATH%"=="" (
-    IF EXIST "C:\Program Files\NVIDIA Corporation\NvToolsExt\lib\x64\nvToolsExt64_1.lib"  (
-        set NVTOOLSEXT_PATH=C:\Program Files\NVIDIA Corporation\NvToolsExt
-    ) ELSE (
-        echo NVTX ^(Visual Studio Extension ^for CUDA^) ^not installed, failing
-        exit /b 1
-    )
-)
-
-IF "%CUDA_PATH_V128%"=="" (
-    IF EXIST "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\bin\nvcc.exe" (
-        set "CUDA_PATH_V128=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8"
-    ) ELSE (
-        echo CUDA 12.8 not found, failing
-        exit /b 1
-    )
-)
-
-IF "%BUILD_VISION%" == "" (
-    set TORCH_CUDA_ARCH_LIST=5.0;6.0;6.1;7.0;7.5;8.0;8.6;9.0;10.0;12.0
-    set TORCH_NVCC_FLAGS=-Xfatbin -compress-all
-) ELSE (
-    set NVCC_FLAGS=-D__CUDA_NO_HALF_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_50,code=sm_50 -gencode=arch=compute_60,code=sm_60 -gencode=arch=compute_70,code=sm_70 -gencode=arch=compute_75,code=sm_75 -gencode=arch=compute_80,code=compute_80 -gencode=arch=compute_86,code=compute_86 -gencode=arch=compute_90,code=compute_90 -gencode=arch=compute_100,code=compute_100 -gencode=arch=compute_120,code=compute_120
-)
-
-set "CUDA_PATH=%CUDA_PATH_V128%"
-set "PATH=%CUDA_PATH_V128%\bin;%PATH%"
-
-:optcheck
-
-call internal\check_opts.bat
-IF ERRORLEVEL 1 goto :eof
-
-if exist "%NIGHTLIES_PYTORCH_ROOT%" cd %NIGHTLIES_PYTORCH_ROOT%\..
-call  %~dp0\internal\copy.bat
-IF ERRORLEVEL 1 goto :eof
-
-call  %~dp0\internal\setup.bat
-IF ERRORLEVEL 1 goto :eof

.ci/pytorch/windows/internal/7z_install.bat

@@ -1,9 +0,0 @@
-@echo off
-
-curl -k https://www.7-zip.org/a/7z1805-x64.exe -O
-if errorlevel 1 exit /b 1
-
-start /wait 7z1805-x64.exe /S
-if errorlevel 1 exit /b 1
-
-set "PATH=%ProgramFiles%\7-Zip;%PATH%"

.ci/pytorch/windows/internal/build_wheels.bat

@@ -1,11 +0,0 @@
-call windows/internal/vc_install_helper.bat
-if errorlevel 1 exit /b 1
-
-call windows/internal/cuda_install.bat
-if errorlevel 1 exit /b 1
-
-call windows/internal/xpu_install.bat
-if errorlevel 1 exit /b 1
-
-call windows/build_pytorch.bat %CUDA_VERSION% %PYTORCH_BUILD_VERSION% %PYTORCH_BUILD_NUMBER%
-if errorlevel 1 exit /b 1

.ci/pytorch/windows/internal/check_deps.bat

@@ -1,80 +0,0 @@
-@echo off
-
-REM Check for necessary components
-echo "Checking dependencies"
-IF NOT "%PROCESSOR_ARCHITECTURE%"=="AMD64" (
-    echo You should use 64 bits Windows to build and run PyTorch
-    exit /b 1
-)
-
-echo "Cmake check"
-IF "%BUILD_VISION%" == "" (
-    where /q cmake.exe
-
-    IF ERRORLEVEL 1 (
-        echo CMake is required to compile PyTorch on Windows
-        exit /b 1
-    )
-)
-
-if not exist "%ProgramFiles(x86)%\Microsoft Visual Studio\Installer\vswhere.exe" (
-    echo Visual Studio %VC_YEAR% C++ BuildTools is required to compile PyTorch on Windows
-    exit /b 1
-)
-
-set VC_VERSION_LOWER=17
-set VC_VERSION_UPPER=18
-if "%VC_YEAR%" == "2019" (
-    set VC_VERSION_LOWER=16
-    set VC_VERSION_UPPER=17
-)
-
-if NOT "%VS15INSTALLDIR%" == "" if exist "%VS15INSTALLDIR%\VC\Auxiliary\Build\vcvarsall.bat" (
-    set "VS15VCVARSALL=%VS15INSTALLDIR%\VC\Auxiliary\Build\vcvarsall.bat"
-    goto vswhere
-)
-
-for /f "usebackq tokens=*" %%i in (`"%ProgramFiles(x86)%\Microsoft Visual Studio\Installer\vswhere.exe" -legacy -products * -version [%VC_VERSION_LOWER%^,%VC_VERSION_UPPER%^) -property installationPath`) do (
-    if exist "%%i" if exist "%%i\VC\Auxiliary\Build\vcvarsall.bat" (
-        set "VS15INSTALLDIR=%%i"
-        set "VS15VCVARSALL=%%i\VC\Auxiliary\Build\vcvarsall.bat"
-        goto vswhere
-    )
-)
-
-:vswhere
-IF "%VS15VCVARSALL%"=="" (
-    echo Visual Studio %VC_YEAR% C++ BuildTools is required to compile PyTorch on Windows
-    exit /b 1
-)
-
-set MSSdk=1
-set DISTUTILS_USE_SDK=1
-
-where /q python.exe
-
-IF ERRORLEVEL 1 (
-    echo Python x64 3.5 or up is required to compile PyTorch on Windows
-    exit /b 1
-)
-
-for /F "usebackq delims=" %%i in (`python -c "import sys; print('{0[0]}{0[1]}'.format(sys.version_info))"`) do (
-    set /a PYVER=%%i
-)
-
-if  %PYVER% LSS 35 (
-    echo Warning: PyTorch for Python 2 under Windows is experimental.
-    echo Python x64 3.5 or up is recommended to compile PyTorch on Windows
-    echo Maybe you can create a virual environment if you have conda installed:
-    echo ^> conda create -n test python=3.6 pyyaml numpy
-    echo ^> activate test
-)
-
-for /F "usebackq delims=" %%i in (`python -c "import struct;print( 8 * struct.calcsize('P'))"`) do (
-    set /a PYSIZE=%%i
-)
-
-if %PYSIZE% NEQ 64 (
-    echo Python x64 3.5 or up is required to compile PyTorch on Windows
-    exit /b 1
-)

.ci/pytorch/windows/internal/check_opts.bat

@@ -1,47 +0,0 @@
-@echo off
-
-REM Check for optional components
-
-where /q ninja.exe
-
-IF NOT ERRORLEVEL 1 (
-    echo Ninja found, using it to speed up builds
-    set CMAKE_GENERATOR=Ninja
-)
-
-IF "%USE_SCCACHE%" == "0" goto sccache_end
-
-where /q clcache.exe
-
-IF NOT ERRORLEVEL 1 (
-    echo clcache found, using it to speed up builds
-    set CC=clcache
-    set CXX=clcache
-)
-
-where /q sccache-cl.exe
-
-IF NOT ERRORLEVEL 1 (
-    echo sccache-cl found, using it to speed up builds
-    set CC=sccache-cl
-    set CXX=sccache-cl
-)
-
-IF "%CC%" == "sccache-cl" IF "%CXX%" == "sccache-cl" goto sccache_end
-
-where /q sccache.exe
-
-IF NOT ERRORLEVEL 1 (
-    echo sccache found, using it to speed up builds
-    set CC=sccache cl
-    set CXX=sccache cl
-)
-
-:sccache_end
-
-IF exist "%MKLProductDir%\mkl\lib\intel64_win" (
-    echo MKL found, adding it to build
-    set "LIB=%MKLProductDir%\mkl\lib\intel64_win;%MKLProductDir%\compiler\lib\intel64_win;%LIB%";
-)
-
-exit /b 0

.ci/pytorch/windows/internal/clean.bat

@@ -1,5 +0,0 @@
-@echo off
-
-cd %MODULE_NAME%
-python setup.py clean
-cd ..

.ci/pytorch/windows/internal/clone.bat

@@ -1,51 +0,0 @@
-@echo off
-
-:: The conda and wheels jobs are separated on Windows, so we don't need to clone again.
-if not exist "%NIGHTLIES_PYTORCH_ROOT%" goto clone_pytorch
-echo "Changing to NIGHTLIES_PYTORCH_ROOT"
-cd "%NIGHTLIES_PYTORCH_ROOT%"
-goto submodule
-
-:clone_pytorch
-
-git clone https://github.com/%PYTORCH_REPO%/%MODULE_NAME%
-
-cd %MODULE_NAME%
-
-IF NOT "%PYTORCH_BRANCH%" == "latest" goto latest_end
-
-:latest_start
-
-if NOT "%NIGHTLIES_DATE%" == "" goto date_end
-
-:date_start
-
-set "DATE_CMD=Get-Date ([System.TimeZoneInfo]::ConvertTimeFromUtc((Get-Date).ToUniversalTime(), [System.TimeZoneInfo]::FindSystemTimeZoneById('Pacific Standard Time'))) -f 'yyyy_MM_dd'"
-set "DATE_COMPACT_CMD=Get-Date ([System.TimeZoneInfo]::ConvertTimeFromUtc((Get-Date).ToUniversalTime(), [System.TimeZoneInfo]::FindSystemTimeZoneById('Pacific Standard Time'))) -f 'yyyyMMdd'"
-
-FOR /F "delims=" %%i IN ('powershell -c "%DATE_CMD%"') DO set NIGHTLIES_DATE=%%i
-FOR /F "delims=" %%i IN ('powershell -c "%DATE_COMPACT_CMD%"') DO set NIGHTLIES_DATE_COMPACT=%%i
-
-:date_end
-
-if "%NIGHTLIES_DATE_COMPACT%" == "" set NIGHTLIES_DATE_COMPACT=%NIGHTLIES_DATE:~0,4%%NIGHTLIES_DATE:~5,2%%NIGHTLIES_DATE:~8,2%
-
-:: Switch to the latest commit by 11:59 yesterday
-echo PYTORCH_BRANCH is set to latest so I will find the last commit
-echo before 0:00 midnight on %NIGHTLIES_DATE%
-set git_date=%NIGHTLIES_DATE:_=-%
-FOR /F "delims=" %%i IN ('git log --before %git_date% -n 1 "--pretty=%%H"') DO set last_commit=%%i
-echo Setting PYTORCH_BRANCH to %last_commit% since that was the last
-echo commit before %NIGHTLIES_DATE%
-set PYTORCH_BRANCH=%last_commit%
-
-:latest_end
-
-IF "%PYTORCH_BRANCH%" == "" set PYTORCH_BRANCH=v%PYTORCH_BUILD_VERSION%
-git checkout %PYTORCH_BRANCH%
-IF ERRORLEVEL 1 git checkout tags/%PYTORCH_BRANCH%
-
-:submodule
-
-git submodule update --init --recursive
-IF ERRORLEVEL 1 exit /b 1

.ci/pytorch/windows/internal/copy.bat

@@ -1,26 +0,0 @@
-copy "%CUDA_PATH%\bin\cusparse*64_*.dll*" pytorch\torch\lib
-copy "%CUDA_PATH%\bin\cublas*64_*.dll*" pytorch\torch\lib
-copy "%CUDA_PATH%\bin\cudart*64_*.dll*" pytorch\torch\lib
-copy "%CUDA_PATH%\bin\curand*64_*.dll*" pytorch\torch\lib
-copy "%CUDA_PATH%\bin\cufft*64_*.dll*" pytorch\torch\lib
-copy "%CUDA_PATH%\bin\cusolver*64_*.dll*" pytorch\torch\lib
-
-copy "%CUDA_PATH%\bin\cudnn*64_*.dll*" pytorch\torch\lib
-copy "%CUDA_PATH%\bin\nvrtc*64_*.dll*" pytorch\torch\lib
-copy "%CUDA_PATH%\extras\CUPTI\lib64\cupti64_*.dll*" pytorch\torch\lib
-
-copy "C:\Program Files\NVIDIA Corporation\NvToolsExt\bin\x64\nvToolsExt64_1.dll*" pytorch\torch\lib
-copy "%CONDA_LIB_PATH%\libiomp*5md.dll" pytorch\torch\lib
-
-:: Should be set in build_pytorch.bat
-copy "%libuv_ROOT%\bin\uv.dll" pytorch\torch\lib
-
-::copy zlib if it exist in windows/system32
-if exist "C:\Windows\System32\zlibwapi.dll" (
-    copy "C:\Windows\System32\zlibwapi.dll"  pytorch\torch\lib
-)
-
-::copy nvJitLink dll is requires for cuda 12+
-if exist "%CUDA_PATH%\bin\nvJitLink_*.dll*" (
-    copy "%CUDA_PATH%\bin\nvJitLink_*.dll*"  pytorch\torch\lib
-)

.ci/pytorch/windows/internal/copy_cpu.bat

@@ -1,3 +0,0 @@
-copy "%CONDA_LIB_PATH%\libiomp*5md.dll" pytorch\torch\lib
-:: Should be set in build_pytorch.bat
-copy "%libuv_ROOT%\bin\uv.dll" pytorch\torch\lib

.ci/pytorch/windows/internal/driver_update.bat

@@ -1,9 +0,0 @@
-set WIN_DRIVER_VN=528.89
-set "DRIVER_DOWNLOAD_LINK=https://ossci-windows.s3.amazonaws.com/%WIN_DRIVER_VN%-data-center-tesla-desktop-winserver-2016-2019-2022-dch-international.exe"
-curl --retry 3 -kL %DRIVER_DOWNLOAD_LINK% --output %WIN_DRIVER_VN%-data-center-tesla-desktop-winserver-2016-2019-2022-dch-international.exe
-if errorlevel 1 exit /b 1
-
-start /wait %WIN_DRIVER_VN%-data-center-tesla-desktop-winserver-2016-2019-2022-dch-international.exe -s -noreboot
-if errorlevel 1 exit /b 1
-
-del %WIN_DRIVER_VN%-data-center-tesla-desktop-winserver-2016-2019-2022-dch-international.exe || ver > NUL