mirror of
https://github.com/pytorch/pytorch.git
synced 2025-10-21 05:34:18 +08:00
Add windows CUDA 12.6 nightly builds (#141805)
Windows AMI was published to prod. This PR adds CUDA 12.6 nightly builds Pull Request resolved: https://github.com/pytorch/pytorch/pull/141805 Approved by: https://github.com/huydhn, https://github.com/Skylion007
This commit is contained in:
atalman
committed by
PyTorch MergeBot
PyTorch MergeBot
parent
9532589b53
commit
0f3f801fc2
@ -245,9 +245,6 @@ def generate_libtorch_matrix(
|
|||||||
arches += ROCM_ARCHES
|
arches += ROCM_ARCHES
|
||||||
elif os == "windows":
|
elif os == "windows":
|
||||||
arches += CUDA_ARCHES
|
arches += CUDA_ARCHES
|
||||||
# skip CUDA 12.6 builds on Windows
|
|
||||||
if "12.6" in arches:
|
|
||||||
arches.remove("12.6")
|
|
||||||
if libtorch_variants is None:
|
if libtorch_variants is None:
|
||||||
libtorch_variants = [
|
libtorch_variants = [
|
||||||
"shared-with-deps",
|
"shared-with-deps",
|
||||||
@ -312,9 +309,6 @@ def generate_wheels_matrix(
|
|||||||
arches += CPU_CXX11_ABI_ARCH + CUDA_ARCHES + ROCM_ARCHES + XPU_ARCHES
|
arches += CPU_CXX11_ABI_ARCH + CUDA_ARCHES + ROCM_ARCHES + XPU_ARCHES
|
||||||
elif os == "windows":
|
elif os == "windows":
|
||||||
arches += CUDA_ARCHES + XPU_ARCHES
|
arches += CUDA_ARCHES + XPU_ARCHES
|
||||||
# skip CUDA 12.6 builds on Windows
|
|
||||||
if "12.6" in arches:
|
|
||||||
arches.remove("12.6")
|
|
||||||
elif os == "linux-aarch64":
|
elif os == "linux-aarch64":
|
||||||
# Only want the one arch as the CPU type is different and
|
# Only want the one arch as the CPU type is different and
|
||||||
# uses different build/test scripts
|
# uses different build/test scripts
|
||||||
|
|||||||
280
.github/workflows/generated-windows-binary-libtorch-debug-nightly.yml
generated
vendored
280
.github/workflows/generated-windows-binary-libtorch-debug-nightly.yml
generated
vendored
@ -878,3 +878,283 @@ jobs:
|
|||||||
conda-pytorchbot-token: ${{ secrets.CONDA_PYTORCHBOT_TOKEN }}
|
conda-pytorchbot-token: ${{ secrets.CONDA_PYTORCHBOT_TOKEN }}
|
||||||
conda-pytorchbot-token-test: ${{ secrets.CONDA_PYTORCHBOT_TOKEN_TEST }}
|
conda-pytorchbot-token-test: ${{ secrets.CONDA_PYTORCHBOT_TOKEN_TEST }}
|
||||||
uses: ./.github/workflows/_binary-upload.yml
|
uses: ./.github/workflows/_binary-upload.yml
|
||||||
|
libtorch-cuda12_6-shared-with-deps-debug-build:
|
||||||
|
if: ${{ github.repository_owner == 'pytorch' }}
|
||||||
|
needs: get-label-type
|
||||||
|
runs-on: "${{ needs.get-label-type.outputs.label-type }}windows.4xlarge"
|
||||||
|
timeout-minutes: 240
|
||||||
|
env:
|
||||||
|
PYTORCH_ROOT: ${{ github.workspace }}/pytorch
|
||||||
|
BUILDER_ROOT: ${{ github.workspace }}/builder
|
||||||
|
PACKAGE_TYPE: libtorch
|
||||||
|
# TODO: This is a legacy variable that we eventually want to get rid of in
|
||||||
|
# favor of GPU_ARCH_VERSION
|
||||||
|
DESIRED_CUDA: cu126
|
||||||
|
GPU_ARCH_VERSION: 12.6
|
||||||
|
GPU_ARCH_TYPE: cuda
|
||||||
|
SKIP_ALL_TESTS: 1
|
||||||
|
LIBTORCH_CONFIG: debug
|
||||||
|
LIBTORCH_VARIANT: shared-with-deps
|
||||||
|
# This is a dummy value for libtorch to work correctly with our batch scripts
|
||||||
|
# without this value pip does not get installed for some reason
|
||||||
|
DESIRED_PYTHON: "3.9"
|
||||||
|
steps:
|
||||||
|
- name: Display EC2 information
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
set -euo pipefail
|
||||||
|
function get_ec2_metadata() {
|
||||||
|
# Pulled from instance metadata endpoint for EC2
|
||||||
|
# see https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/instancedata-data-retrieval.html
|
||||||
|
category=$1
|
||||||
|
curl -H "X-aws-ec2-metadata-token: $(curl -s -X PUT "http://169.254.169.254/latest/api/token" -H "X-aws-ec2-metadata-token-ttl-seconds: 30")" -fsSL "http://169.254.169.254/latest/meta-data/${category}"
|
||||||
|
}
|
||||||
|
echo "ami-id: $(get_ec2_metadata ami-id)"
|
||||||
|
echo "instance-id: $(get_ec2_metadata instance-id)"
|
||||||
|
echo "instance-type: $(get_ec2_metadata instance-type)"
|
||||||
|
echo "system info $(uname -a)"
|
||||||
|
- name: "[FB EMPLOYEES] Enable SSH (Click me for login details)"
|
||||||
|
uses: pytorch/test-infra/.github/actions/setup-ssh@main
|
||||||
|
continue-on-error: true
|
||||||
|
with:
|
||||||
|
github-secret: ${{ secrets.GITHUB_TOKEN }}
|
||||||
|
- name: Enable git long paths and symlinks on Windows and disable fsmonitor daemon
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
git config --global core.longpaths true
|
||||||
|
git config --global core.symlinks true
|
||||||
|
|
||||||
|
# https://git-scm.com/docs/git-fsmonitor--daemon. The daemon could lock
|
||||||
|
# the directory on Windows and prevent GHA from checking out as reported
|
||||||
|
# in https://github.com/actions/checkout/issues/1018
|
||||||
|
git config --global core.fsmonitor false
|
||||||
|
# Needed for binary builds, see: https://github.com/pytorch/pytorch/issues/73339#issuecomment-1058981560
|
||||||
|
- name: Enable long paths on Windows
|
||||||
|
shell: powershell
|
||||||
|
run: |
|
||||||
|
Set-ItemProperty -Path "HKLM:\\SYSTEM\CurrentControlSet\Control\FileSystem" -Name "LongPathsEnabled" -Value 1
|
||||||
|
# Since it's just a defensive command, the workflow should continue even the command fails. This step can be
|
||||||
|
# removed once Windows Defender is removed from the AMI
|
||||||
|
- name: Disables Windows Defender scheduled and real-time scanning for files in directories used by PyTorch
|
||||||
|
continue-on-error: true
|
||||||
|
shell: powershell
|
||||||
|
run: |
|
||||||
|
Add-MpPreference -ExclusionPath $(Get-Location).tostring(),$Env:TEMP -ErrorAction Ignore
|
||||||
|
# Let's both exclude the path and disable Windows Defender completely just to be sure
|
||||||
|
# that it doesn't interfere
|
||||||
|
Set-MpPreference -DisableRealtimeMonitoring $True -ErrorAction Ignore
|
||||||
|
# NOTE: These environment variables are put here so that they can be applied on every job equally
|
||||||
|
# They are also here because setting them at a workflow level doesn't give us access to the
|
||||||
|
# runner.temp variable, which we need.
|
||||||
|
- name: Populate binary env
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
echo "BINARY_ENV_FILE=${RUNNER_TEMP}/env" >> "${GITHUB_ENV}"
|
||||||
|
echo "PYTORCH_FINAL_PACKAGE_DIR=${RUNNER_TEMP}/artifacts" >> "${GITHUB_ENV}"
|
||||||
|
echo "WIN_PACKAGE_WORK_DIR=${RUNNER_TEMP}"
|
||||||
|
- name: Checkout PyTorch
|
||||||
|
uses: malfet/checkout@silent-checkout
|
||||||
|
with:
|
||||||
|
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
|
||||||
|
submodules: recursive
|
||||||
|
path: pytorch
|
||||||
|
quiet-checkout: true
|
||||||
|
- name: Clean PyTorch checkout
|
||||||
|
run: |
|
||||||
|
# Remove any artifacts from the previous checkouts
|
||||||
|
git clean -fxd
|
||||||
|
working-directory: pytorch
|
||||||
|
- name: Checkout pytorch/builder
|
||||||
|
uses: malfet/checkout@silent-checkout
|
||||||
|
with:
|
||||||
|
ref: main
|
||||||
|
submodules: recursive
|
||||||
|
repository: pytorch/builder
|
||||||
|
path: builder
|
||||||
|
quiet-checkout: true
|
||||||
|
- name: Clean pytorch/builder checkout
|
||||||
|
run: |
|
||||||
|
# Remove any artifacts from the previous checkouts
|
||||||
|
git clean -fxd
|
||||||
|
working-directory: builder
|
||||||
|
- name: Populate binary env
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
"${PYTORCH_ROOT}/.circleci/scripts/binary_populate_env.sh"
|
||||||
|
- name: Build PyTorch binary
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
"${PYTORCH_ROOT}/.circleci/scripts/binary_windows_build.sh"
|
||||||
|
- uses: actions/upload-artifact@v4.4.0
|
||||||
|
if: always()
|
||||||
|
with:
|
||||||
|
name: libtorch-cuda12_6-shared-with-deps-debug
|
||||||
|
retention-days: 14
|
||||||
|
if-no-files-found: error
|
||||||
|
path: "${{ env.PYTORCH_FINAL_PACKAGE_DIR }}"
|
||||||
|
- name: Wait until all sessions have drained
|
||||||
|
shell: powershell
|
||||||
|
working-directory: pytorch
|
||||||
|
if: always()
|
||||||
|
timeout-minutes: 120
|
||||||
|
run: |
|
||||||
|
.github\scripts\wait_for_ssh_to_drain.ps1
|
||||||
|
- name: Kill active ssh sessions if still around (Useful if workflow was cancelled)
|
||||||
|
shell: powershell
|
||||||
|
working-directory: pytorch
|
||||||
|
if: always()
|
||||||
|
run: |
|
||||||
|
.github\scripts\kill_active_ssh_sessions.ps1
|
||||||
|
libtorch-cuda12_6-shared-with-deps-debug-test: # Testing
|
||||||
|
if: ${{ github.repository_owner == 'pytorch' }}
|
||||||
|
needs:
|
||||||
|
- libtorch-cuda12_6-shared-with-deps-debug-build
|
||||||
|
- get-label-type
|
||||||
|
runs-on: "${{ needs.get-label-type.outputs.label-type }}windows.g4dn.xlarge"
|
||||||
|
timeout-minutes: 240
|
||||||
|
env:
|
||||||
|
PYTORCH_ROOT: ${{ github.workspace }}/pytorch
|
||||||
|
BUILDER_ROOT: ${{ github.workspace }}/builder
|
||||||
|
PACKAGE_TYPE: libtorch
|
||||||
|
# TODO: This is a legacy variable that we eventually want to get rid of in
|
||||||
|
# favor of GPU_ARCH_VERSION
|
||||||
|
DESIRED_CUDA: cu126
|
||||||
|
GPU_ARCH_VERSION: 12.6
|
||||||
|
GPU_ARCH_TYPE: cuda
|
||||||
|
SKIP_ALL_TESTS: 1
|
||||||
|
LIBTORCH_CONFIG: debug
|
||||||
|
LIBTORCH_VARIANT: shared-with-deps
|
||||||
|
# This is a dummy value for libtorch to work correctly with our batch scripts
|
||||||
|
# without this value pip does not get installed for some reason
|
||||||
|
DESIRED_PYTHON: "3.9"
|
||||||
|
steps:
|
||||||
|
- name: Display EC2 information
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
set -euo pipefail
|
||||||
|
function get_ec2_metadata() {
|
||||||
|
# Pulled from instance metadata endpoint for EC2
|
||||||
|
# see https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/instancedata-data-retrieval.html
|
||||||
|
category=$1
|
||||||
|
curl -H "X-aws-ec2-metadata-token: $(curl -s -X PUT "http://169.254.169.254/latest/api/token" -H "X-aws-ec2-metadata-token-ttl-seconds: 30")" -fsSL "http://169.254.169.254/latest/meta-data/${category}"
|
||||||
|
}
|
||||||
|
echo "ami-id: $(get_ec2_metadata ami-id)"
|
||||||
|
echo "instance-id: $(get_ec2_metadata instance-id)"
|
||||||
|
echo "instance-type: $(get_ec2_metadata instance-type)"
|
||||||
|
echo "system info $(uname -a)"
|
||||||
|
- name: "[FB EMPLOYEES] Enable SSH (Click me for login details)"
|
||||||
|
uses: pytorch/test-infra/.github/actions/setup-ssh@main
|
||||||
|
continue-on-error: true
|
||||||
|
with:
|
||||||
|
github-secret: ${{ secrets.GITHUB_TOKEN }}
|
||||||
|
- name: Enable git long paths and symlinks on Windows and disable fsmonitor daemon
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
git config --global core.longpaths true
|
||||||
|
git config --global core.symlinks true
|
||||||
|
|
||||||
|
# https://git-scm.com/docs/git-fsmonitor--daemon. The daemon could lock
|
||||||
|
# the directory on Windows and prevent GHA from checking out as reported
|
||||||
|
# in https://github.com/actions/checkout/issues/1018
|
||||||
|
git config --global core.fsmonitor false
|
||||||
|
# Needed for binary builds, see: https://github.com/pytorch/pytorch/issues/73339#issuecomment-1058981560
|
||||||
|
- name: Enable long paths on Windows
|
||||||
|
shell: powershell
|
||||||
|
run: |
|
||||||
|
Set-ItemProperty -Path "HKLM:\\SYSTEM\CurrentControlSet\Control\FileSystem" -Name "LongPathsEnabled" -Value 1
|
||||||
|
# Since it's just a defensive command, the workflow should continue even the command fails. This step can be
|
||||||
|
# removed once Windows Defender is removed from the AMI
|
||||||
|
- name: Disables Windows Defender scheduled and real-time scanning for files in directories used by PyTorch
|
||||||
|
continue-on-error: true
|
||||||
|
shell: powershell
|
||||||
|
run: |
|
||||||
|
Add-MpPreference -ExclusionPath $(Get-Location).tostring(),$Env:TEMP -ErrorAction Ignore
|
||||||
|
# Let's both exclude the path and disable Windows Defender completely just to be sure
|
||||||
|
# that it doesn't interfere
|
||||||
|
Set-MpPreference -DisableRealtimeMonitoring $True -ErrorAction Ignore
|
||||||
|
# NOTE: These environment variables are put here so that they can be applied on every job equally
|
||||||
|
# They are also here because setting them at a workflow level doesn't give us access to the
|
||||||
|
# runner.temp variable, which we need.
|
||||||
|
- name: Populate binary env
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
echo "BINARY_ENV_FILE=${RUNNER_TEMP}/env" >> "${GITHUB_ENV}"
|
||||||
|
echo "PYTORCH_FINAL_PACKAGE_DIR=${RUNNER_TEMP}/artifacts" >> "${GITHUB_ENV}"
|
||||||
|
echo "WIN_PACKAGE_WORK_DIR=${RUNNER_TEMP}"
|
||||||
|
- uses: actions/download-artifact@v4.1.7
|
||||||
|
name: Download Build Artifacts
|
||||||
|
with:
|
||||||
|
name: libtorch-cuda12_6-shared-with-deps-debug
|
||||||
|
path: "${{ env.PYTORCH_FINAL_PACKAGE_DIR }}"
|
||||||
|
- name: Checkout PyTorch
|
||||||
|
uses: malfet/checkout@silent-checkout
|
||||||
|
with:
|
||||||
|
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
|
||||||
|
submodules: recursive
|
||||||
|
path: pytorch
|
||||||
|
quiet-checkout: true
|
||||||
|
- name: Clean PyTorch checkout
|
||||||
|
run: |
|
||||||
|
# Remove any artifacts from the previous checkouts
|
||||||
|
git clean -fxd
|
||||||
|
working-directory: pytorch
|
||||||
|
- name: Checkout pytorch/builder
|
||||||
|
uses: malfet/checkout@silent-checkout
|
||||||
|
with:
|
||||||
|
ref: main
|
||||||
|
submodules: recursive
|
||||||
|
repository: pytorch/builder
|
||||||
|
path: builder
|
||||||
|
quiet-checkout: true
|
||||||
|
- name: Clean pytorch/builder checkout
|
||||||
|
run: |
|
||||||
|
# Remove any artifacts from the previous checkouts
|
||||||
|
git clean -fxd
|
||||||
|
working-directory: builder
|
||||||
|
- name: Populate binary env
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
"${PYTORCH_ROOT}/.circleci/scripts/binary_populate_env.sh"
|
||||||
|
- name: Test PyTorch binary
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
"${PYTORCH_ROOT}/.circleci/scripts/binary_windows_test.sh"
|
||||||
|
- name: Wait until all sessions have drained
|
||||||
|
shell: powershell
|
||||||
|
working-directory: pytorch
|
||||||
|
if: always()
|
||||||
|
timeout-minutes: 120
|
||||||
|
run: |
|
||||||
|
.github\scripts\wait_for_ssh_to_drain.ps1
|
||||||
|
- name: Kill active ssh sessions if still around (Useful if workflow was cancelled)
|
||||||
|
shell: powershell
|
||||||
|
working-directory: pytorch
|
||||||
|
if: always()
|
||||||
|
run: |
|
||||||
|
.github\scripts\kill_active_ssh_sessions.ps1
|
||||||
|
libtorch-cuda12_6-shared-with-deps-debug-upload: # Uploading
|
||||||
|
if: ${{ github.repository_owner == 'pytorch' }}
|
||||||
|
permissions:
|
||||||
|
id-token: write
|
||||||
|
contents: read
|
||||||
|
needs: libtorch-cuda12_6-shared-with-deps-debug-test
|
||||||
|
with:
|
||||||
|
PYTORCH_ROOT: ${{ github.workspace }}/pytorch
|
||||||
|
BUILDER_ROOT: ${{ github.workspace }}/builder
|
||||||
|
PACKAGE_TYPE: libtorch
|
||||||
|
# TODO: This is a legacy variable that we eventually want to get rid of in
|
||||||
|
# favor of GPU_ARCH_VERSION
|
||||||
|
DESIRED_CUDA: cu126
|
||||||
|
GPU_ARCH_VERSION: 12.6
|
||||||
|
GPU_ARCH_TYPE: cuda
|
||||||
|
LIBTORCH_CONFIG: debug
|
||||||
|
LIBTORCH_VARIANT: shared-with-deps
|
||||||
|
# This is a dummy value for libtorch to work correctly with our batch scripts
|
||||||
|
# without this value pip does not get installed for some reason
|
||||||
|
DESIRED_PYTHON: "3.9"
|
||||||
|
build_name: libtorch-cuda12_6-shared-with-deps-debug
|
||||||
|
secrets:
|
||||||
|
github-token: ${{ secrets.GITHUB_TOKEN }}
|
||||||
|
conda-pytorchbot-token: ${{ secrets.CONDA_PYTORCHBOT_TOKEN }}
|
||||||
|
conda-pytorchbot-token-test: ${{ secrets.CONDA_PYTORCHBOT_TOKEN_TEST }}
|
||||||
|
uses: ./.github/workflows/_binary-upload.yml
|
||||||
|
|||||||
280
.github/workflows/generated-windows-binary-libtorch-release-nightly.yml
generated
vendored
280
.github/workflows/generated-windows-binary-libtorch-release-nightly.yml
generated
vendored
@ -878,3 +878,283 @@ jobs:
|
|||||||
conda-pytorchbot-token: ${{ secrets.CONDA_PYTORCHBOT_TOKEN }}
|
conda-pytorchbot-token: ${{ secrets.CONDA_PYTORCHBOT_TOKEN }}
|
||||||
conda-pytorchbot-token-test: ${{ secrets.CONDA_PYTORCHBOT_TOKEN_TEST }}
|
conda-pytorchbot-token-test: ${{ secrets.CONDA_PYTORCHBOT_TOKEN_TEST }}
|
||||||
uses: ./.github/workflows/_binary-upload.yml
|
uses: ./.github/workflows/_binary-upload.yml
|
||||||
|
libtorch-cuda12_6-shared-with-deps-release-build:
|
||||||
|
if: ${{ github.repository_owner == 'pytorch' }}
|
||||||
|
needs: get-label-type
|
||||||
|
runs-on: "${{ needs.get-label-type.outputs.label-type }}windows.4xlarge"
|
||||||
|
timeout-minutes: 240
|
||||||
|
env:
|
||||||
|
PYTORCH_ROOT: ${{ github.workspace }}/pytorch
|
||||||
|
BUILDER_ROOT: ${{ github.workspace }}/builder
|
||||||
|
PACKAGE_TYPE: libtorch
|
||||||
|
# TODO: This is a legacy variable that we eventually want to get rid of in
|
||||||
|
# favor of GPU_ARCH_VERSION
|
||||||
|
DESIRED_CUDA: cu126
|
||||||
|
GPU_ARCH_VERSION: 12.6
|
||||||
|
GPU_ARCH_TYPE: cuda
|
||||||
|
SKIP_ALL_TESTS: 1
|
||||||
|
LIBTORCH_CONFIG: release
|
||||||
|
LIBTORCH_VARIANT: shared-with-deps
|
||||||
|
# This is a dummy value for libtorch to work correctly with our batch scripts
|
||||||
|
# without this value pip does not get installed for some reason
|
||||||
|
DESIRED_PYTHON: "3.9"
|
||||||
|
steps:
|
||||||
|
- name: Display EC2 information
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
set -euo pipefail
|
||||||
|
function get_ec2_metadata() {
|
||||||
|
# Pulled from instance metadata endpoint for EC2
|
||||||
|
# see https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/instancedata-data-retrieval.html
|
||||||
|
category=$1
|
||||||
|
curl -H "X-aws-ec2-metadata-token: $(curl -s -X PUT "http://169.254.169.254/latest/api/token" -H "X-aws-ec2-metadata-token-ttl-seconds: 30")" -fsSL "http://169.254.169.254/latest/meta-data/${category}"
|
||||||
|
}
|
||||||
|
echo "ami-id: $(get_ec2_metadata ami-id)"
|
||||||
|
echo "instance-id: $(get_ec2_metadata instance-id)"
|
||||||
|
echo "instance-type: $(get_ec2_metadata instance-type)"
|
||||||
|
echo "system info $(uname -a)"
|
||||||
|
- name: "[FB EMPLOYEES] Enable SSH (Click me for login details)"
|
||||||
|
uses: pytorch/test-infra/.github/actions/setup-ssh@main
|
||||||
|
continue-on-error: true
|
||||||
|
with:
|
||||||
|
github-secret: ${{ secrets.GITHUB_TOKEN }}
|
||||||
|
- name: Enable git long paths and symlinks on Windows and disable fsmonitor daemon
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
git config --global core.longpaths true
|
||||||
|
git config --global core.symlinks true
|
||||||
|
|
||||||
|
# https://git-scm.com/docs/git-fsmonitor--daemon. The daemon could lock
|
||||||
|
# the directory on Windows and prevent GHA from checking out as reported
|
||||||
|
# in https://github.com/actions/checkout/issues/1018
|
||||||
|
git config --global core.fsmonitor false
|
||||||
|
# Needed for binary builds, see: https://github.com/pytorch/pytorch/issues/73339#issuecomment-1058981560
|
||||||
|
- name: Enable long paths on Windows
|
||||||
|
shell: powershell
|
||||||
|
run: |
|
||||||
|
Set-ItemProperty -Path "HKLM:\\SYSTEM\CurrentControlSet\Control\FileSystem" -Name "LongPathsEnabled" -Value 1
|
||||||
|
# Since it's just a defensive command, the workflow should continue even the command fails. This step can be
|
||||||
|
# removed once Windows Defender is removed from the AMI
|
||||||
|
- name: Disables Windows Defender scheduled and real-time scanning for files in directories used by PyTorch
|
||||||
|
continue-on-error: true
|
||||||
|
shell: powershell
|
||||||
|
run: |
|
||||||
|
Add-MpPreference -ExclusionPath $(Get-Location).tostring(),$Env:TEMP -ErrorAction Ignore
|
||||||
|
# Let's both exclude the path and disable Windows Defender completely just to be sure
|
||||||
|
# that it doesn't interfere
|
||||||
|
Set-MpPreference -DisableRealtimeMonitoring $True -ErrorAction Ignore
|
||||||
|
# NOTE: These environment variables are put here so that they can be applied on every job equally
|
||||||
|
# They are also here because setting them at a workflow level doesn't give us access to the
|
||||||
|
# runner.temp variable, which we need.
|
||||||
|
- name: Populate binary env
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
echo "BINARY_ENV_FILE=${RUNNER_TEMP}/env" >> "${GITHUB_ENV}"
|
||||||
|
echo "PYTORCH_FINAL_PACKAGE_DIR=${RUNNER_TEMP}/artifacts" >> "${GITHUB_ENV}"
|
||||||
|
echo "WIN_PACKAGE_WORK_DIR=${RUNNER_TEMP}"
|
||||||
|
- name: Checkout PyTorch
|
||||||
|
uses: malfet/checkout@silent-checkout
|
||||||
|
with:
|
||||||
|
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
|
||||||
|
submodules: recursive
|
||||||
|
path: pytorch
|
||||||
|
quiet-checkout: true
|
||||||
|
- name: Clean PyTorch checkout
|
||||||
|
run: |
|
||||||
|
# Remove any artifacts from the previous checkouts
|
||||||
|
git clean -fxd
|
||||||
|
working-directory: pytorch
|
||||||
|
- name: Checkout pytorch/builder
|
||||||
|
uses: malfet/checkout@silent-checkout
|
||||||
|
with:
|
||||||
|
ref: main
|
||||||
|
submodules: recursive
|
||||||
|
repository: pytorch/builder
|
||||||
|
path: builder
|
||||||
|
quiet-checkout: true
|
||||||
|
- name: Clean pytorch/builder checkout
|
||||||
|
run: |
|
||||||
|
# Remove any artifacts from the previous checkouts
|
||||||
|
git clean -fxd
|
||||||
|
working-directory: builder
|
||||||
|
- name: Populate binary env
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
"${PYTORCH_ROOT}/.circleci/scripts/binary_populate_env.sh"
|
||||||
|
- name: Build PyTorch binary
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
"${PYTORCH_ROOT}/.circleci/scripts/binary_windows_build.sh"
|
||||||
|
- uses: actions/upload-artifact@v4.4.0
|
||||||
|
if: always()
|
||||||
|
with:
|
||||||
|
name: libtorch-cuda12_6-shared-with-deps-release
|
||||||
|
retention-days: 14
|
||||||
|
if-no-files-found: error
|
||||||
|
path: "${{ env.PYTORCH_FINAL_PACKAGE_DIR }}"
|
||||||
|
- name: Wait until all sessions have drained
|
||||||
|
shell: powershell
|
||||||
|
working-directory: pytorch
|
||||||
|
if: always()
|
||||||
|
timeout-minutes: 120
|
||||||
|
run: |
|
||||||
|
.github\scripts\wait_for_ssh_to_drain.ps1
|
||||||
|
- name: Kill active ssh sessions if still around (Useful if workflow was cancelled)
|
||||||
|
shell: powershell
|
||||||
|
working-directory: pytorch
|
||||||
|
if: always()
|
||||||
|
run: |
|
||||||
|
.github\scripts\kill_active_ssh_sessions.ps1
|
||||||
|
libtorch-cuda12_6-shared-with-deps-release-test: # Testing
|
||||||
|
if: ${{ github.repository_owner == 'pytorch' }}
|
||||||
|
needs:
|
||||||
|
- libtorch-cuda12_6-shared-with-deps-release-build
|
||||||
|
- get-label-type
|
||||||
|
runs-on: "${{ needs.get-label-type.outputs.label-type }}windows.g4dn.xlarge"
|
||||||
|
timeout-minutes: 240
|
||||||
|
env:
|
||||||
|
PYTORCH_ROOT: ${{ github.workspace }}/pytorch
|
||||||
|
BUILDER_ROOT: ${{ github.workspace }}/builder
|
||||||
|
PACKAGE_TYPE: libtorch
|
||||||
|
# TODO: This is a legacy variable that we eventually want to get rid of in
|
||||||
|
# favor of GPU_ARCH_VERSION
|
||||||
|
DESIRED_CUDA: cu126
|
||||||
|
GPU_ARCH_VERSION: 12.6
|
||||||
|
GPU_ARCH_TYPE: cuda
|
||||||
|
SKIP_ALL_TESTS: 1
|
||||||
|
LIBTORCH_CONFIG: release
|
||||||
|
LIBTORCH_VARIANT: shared-with-deps
|
||||||
|
# This is a dummy value for libtorch to work correctly with our batch scripts
|
||||||
|
# without this value pip does not get installed for some reason
|
||||||
|
DESIRED_PYTHON: "3.9"
|
||||||
|
steps:
|
||||||
|
- name: Display EC2 information
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
set -euo pipefail
|
||||||
|
function get_ec2_metadata() {
|
||||||
|
# Pulled from instance metadata endpoint for EC2
|
||||||
|
# see https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/instancedata-data-retrieval.html
|
||||||
|
category=$1
|
||||||
|
curl -H "X-aws-ec2-metadata-token: $(curl -s -X PUT "http://169.254.169.254/latest/api/token" -H "X-aws-ec2-metadata-token-ttl-seconds: 30")" -fsSL "http://169.254.169.254/latest/meta-data/${category}"
|
||||||
|
}
|
||||||
|
echo "ami-id: $(get_ec2_metadata ami-id)"
|
||||||
|
echo "instance-id: $(get_ec2_metadata instance-id)"
|
||||||
|
echo "instance-type: $(get_ec2_metadata instance-type)"
|
||||||
|
echo "system info $(uname -a)"
|
||||||
|
- name: "[FB EMPLOYEES] Enable SSH (Click me for login details)"
|
||||||
|
uses: pytorch/test-infra/.github/actions/setup-ssh@main
|
||||||
|
continue-on-error: true
|
||||||
|
with:
|
||||||
|
github-secret: ${{ secrets.GITHUB_TOKEN }}
|
||||||
|
- name: Enable git long paths and symlinks on Windows and disable fsmonitor daemon
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
git config --global core.longpaths true
|
||||||
|
git config --global core.symlinks true
|
||||||
|
|
||||||
|
# https://git-scm.com/docs/git-fsmonitor--daemon. The daemon could lock
|
||||||
|
# the directory on Windows and prevent GHA from checking out as reported
|
||||||
|
# in https://github.com/actions/checkout/issues/1018
|
||||||
|
git config --global core.fsmonitor false
|
||||||
|
# Needed for binary builds, see: https://github.com/pytorch/pytorch/issues/73339#issuecomment-1058981560
|
||||||
|
- name: Enable long paths on Windows
|
||||||
|
shell: powershell
|
||||||
|
run: |
|
||||||
|
Set-ItemProperty -Path "HKLM:\\SYSTEM\CurrentControlSet\Control\FileSystem" -Name "LongPathsEnabled" -Value 1
|
||||||
|
# Since it's just a defensive command, the workflow should continue even the command fails. This step can be
|
||||||
|
# removed once Windows Defender is removed from the AMI
|
||||||
|
- name: Disables Windows Defender scheduled and real-time scanning for files in directories used by PyTorch
|
||||||
|
continue-on-error: true
|
||||||
|
shell: powershell
|
||||||
|
run: |
|
||||||
|
Add-MpPreference -ExclusionPath $(Get-Location).tostring(),$Env:TEMP -ErrorAction Ignore
|
||||||
|
# Let's both exclude the path and disable Windows Defender completely just to be sure
|
||||||
|
# that it doesn't interfere
|
||||||
|
Set-MpPreference -DisableRealtimeMonitoring $True -ErrorAction Ignore
|
||||||
|
# NOTE: These environment variables are put here so that they can be applied on every job equally
|
||||||
|
# They are also here because setting them at a workflow level doesn't give us access to the
|
||||||
|
# runner.temp variable, which we need.
|
||||||
|
- name: Populate binary env
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
echo "BINARY_ENV_FILE=${RUNNER_TEMP}/env" >> "${GITHUB_ENV}"
|
||||||
|
echo "PYTORCH_FINAL_PACKAGE_DIR=${RUNNER_TEMP}/artifacts" >> "${GITHUB_ENV}"
|
||||||
|
echo "WIN_PACKAGE_WORK_DIR=${RUNNER_TEMP}"
|
||||||
|
- uses: actions/download-artifact@v4.1.7
|
||||||
|
name: Download Build Artifacts
|
||||||
|
with:
|
||||||
|
name: libtorch-cuda12_6-shared-with-deps-release
|
||||||
|
path: "${{ env.PYTORCH_FINAL_PACKAGE_DIR }}"
|
||||||
|
- name: Checkout PyTorch
|
||||||
|
uses: malfet/checkout@silent-checkout
|
||||||
|
with:
|
||||||
|
ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
|
||||||
|
submodules: recursive
|
||||||
|
path: pytorch
|
||||||
|
quiet-checkout: true
|
||||||
|
- name: Clean PyTorch checkout
|
||||||
|
run: |
|
||||||
|
# Remove any artifacts from the previous checkouts
|
||||||
|
git clean -fxd
|
||||||
|
working-directory: pytorch
|
||||||
|
- name: Checkout pytorch/builder
|
||||||
|
uses: malfet/checkout@silent-checkout
|
||||||
|
with:
|
||||||
|
ref: main
|
||||||
|
submodules: recursive
|
||||||
|
repository: pytorch/builder
|
||||||
|
path: builder
|
||||||
|
quiet-checkout: true
|
||||||
|
- name: Clean pytorch/builder checkout
|
||||||
|
run: |
|
||||||
|
# Remove any artifacts from the previous checkouts
|
||||||
|
git clean -fxd
|
||||||
|
working-directory: builder
|
||||||
|
- name: Populate binary env
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
"${PYTORCH_ROOT}/.circleci/scripts/binary_populate_env.sh"
|
||||||
|
- name: Test PyTorch binary
|
||||||
|
shell: bash
|
||||||
|
run: |
|
||||||
|
"${PYTORCH_ROOT}/.circleci/scripts/binary_windows_test.sh"
|
||||||
|
- name: Wait until all sessions have drained
|
||||||
|
shell: powershell
|
||||||
|
working-directory: pytorch
|
||||||
|
if: always()
|
||||||
|
timeout-minutes: 120
|
||||||
|
run: |
|
||||||
|
.github\scripts\wait_for_ssh_to_drain.ps1
|
||||||
|
- name: Kill active ssh sessions if still around (Useful if workflow was cancelled)
|
||||||
|
shell: powershell
|
||||||
|
working-directory: pytorch
|
||||||
|
if: always()
|
||||||
|
run: |
|
||||||
|
.github\scripts\kill_active_ssh_sessions.ps1
|
||||||
|
libtorch-cuda12_6-shared-with-deps-release-upload: # Uploading
|
||||||
|
if: ${{ github.repository_owner == 'pytorch' }}
|
||||||
|
permissions:
|
||||||
|
id-token: write
|
||||||
|
contents: read
|
||||||
|
needs: libtorch-cuda12_6-shared-with-deps-release-test
|
||||||
|
with:
|
||||||
|
PYTORCH_ROOT: ${{ github.workspace }}/pytorch
|
||||||
|
BUILDER_ROOT: ${{ github.workspace }}/builder
|
||||||
|
PACKAGE_TYPE: libtorch
|
||||||
|
# TODO: This is a legacy variable that we eventually want to get rid of in
|
||||||
|
# favor of GPU_ARCH_VERSION
|
||||||
|
DESIRED_CUDA: cu126
|
||||||
|
GPU_ARCH_VERSION: 12.6
|
||||||
|
GPU_ARCH_TYPE: cuda
|
||||||
|
LIBTORCH_CONFIG: release
|
||||||
|
LIBTORCH_VARIANT: shared-with-deps
|
||||||
|
# This is a dummy value for libtorch to work correctly with our batch scripts
|
||||||
|
# without this value pip does not get installed for some reason
|
||||||
|
DESIRED_PYTHON: "3.9"
|
||||||
|
build_name: libtorch-cuda12_6-shared-with-deps-release
|
||||||
|
secrets:
|
||||||
|
github-token: ${{ secrets.GITHUB_TOKEN }}
|
||||||
|
conda-pytorchbot-token: ${{ secrets.CONDA_PYTORCHBOT_TOKEN }}
|
||||||
|
conda-pytorchbot-token-test: ${{ secrets.CONDA_PYTORCHBOT_TOKEN_TEST }}
|
||||||
|
uses: ./.github/workflows/_binary-upload.yml
|
||||||
|
|||||||
1345
.github/workflows/generated-windows-binary-wheel-nightly.yml
generated
vendored
1345
.github/workflows/generated-windows-binary-wheel-nightly.yml
generated
vendored
File diff suppressed because it is too large
Load Diff
Reference in New Issue
Block a user