From a85708c37609964dcc1381a2f930fe37e17e94e3 Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Fri, 10 May 2019 13:34:38 -0700 Subject: [PATCH 01/26] Create azure-pipelines.yml --- azure-pipelines.yml | 39 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+) create mode 100644 azure-pipelines.yml diff --git a/azure-pipelines.yml b/azure-pipelines.yml new file mode 100644 index 000000000000..b55ce6fbf6ba --- /dev/null +++ b/azure-pipelines.yml @@ -0,0 +1,39 @@ +# Starter pipeline +# Start with a minimal pipeline that you can customize to build and deploy your code. +# Add steps that build, run tests, deploy, and more: +# https://aka.ms/yaml + +trigger: +- master + +jobs: + - job: + pool: "Default" + displayName: "Build and Test" + container: + image: 308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/pytorch-linux-trusty-py3.5:300 + endpoint: aws_docker + options: -u 0 + timeoutInMinutes: 120 + steps: + - checkout: self + submodules: "true" + - bash: | + set +x + sudo chown -R azureuser_azpcontainer:azureuser_azpcontainer /usr/local + sudo chown -R azureuser_azpcontainer:azureuser_azpcontainer /opt/python + .jenkins/pytorch/build.sh + env: + BUILD_ENVIRONMENT: pytorch-linux-trusty-py3.5-build + SCCACHE_BUCKET: "ossci-compiler-cache-circleci-v2" + AWS_ACCESS_KEY_ID: $(SCCACHE_ACCESS_KEY_ID) + AWS_SECRET_ACCESS_KEY: $(SCCACHE_SECRET) + displayName: 'Build' + - bash: | + set +x + set -e + export PATH="/home/azureuser_azpcontainer/.local/bin:$PATH" + .jenkins/pytorch/test.sh + env: + BUILD_ENVIRONMENT: pytorch-linux-trusty-py3.5-test + displayName: 'Test' From ef9ca9a0428e6c1f2a6186c0d42702107d1942c1 Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Fri, 10 May 2019 17:42:54 -0700 Subject: [PATCH 02/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 40 +++++++++++++++++++++++++++++++++++++--- 1 file changed, 37 insertions(+), 3 deletions(-) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index b55ce6fbf6ba..7f0a589a8072 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -8,10 +8,12 @@ trigger: jobs: - job: - pool: "Default" + pool: + name: "Default" + demands: "gpu" displayName: "Build and Test" container: - image: 308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/pytorch-linux-trusty-py3.5:300 + image: 308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/pytorch-linux-xenial-cuda9-cudnn7-py3:300 endpoint: aws_docker options: -u 0 timeoutInMinutes: 120 @@ -24,7 +26,7 @@ jobs: sudo chown -R azureuser_azpcontainer:azureuser_azpcontainer /opt/python .jenkins/pytorch/build.sh env: - BUILD_ENVIRONMENT: pytorch-linux-trusty-py3.5-build + BUILD_ENVIRONMENT: pytorch-linux-xenial-cuda9-cudnn7-py3-build SCCACHE_BUCKET: "ossci-compiler-cache-circleci-v2" AWS_ACCESS_KEY_ID: $(SCCACHE_ACCESS_KEY_ID) AWS_SECRET_ACCESS_KEY: $(SCCACHE_SECRET) @@ -37,3 +39,35 @@ jobs: env: BUILD_ENVIRONMENT: pytorch-linux-trusty-py3.5-test displayName: 'Test' + +# jobs: +# - job: +# pool: "Default" +# displayName: "Build and Test" +# container: +# image: 308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/pytorch-linux-trusty-py3.5:300 +# endpoint: aws_docker +# options: -u 0 +# timeoutInMinutes: 120 +# steps: +# - checkout: self +# submodules: "true" +# - bash: | +# set +x +# sudo chown -R azureuser_azpcontainer:azureuser_azpcontainer /usr/local +# sudo chown -R azureuser_azpcontainer:azureuser_azpcontainer /opt/python +# .jenkins/pytorch/build.sh +# env: +# BUILD_ENVIRONMENT: pytorch-linux-trusty-py3.5-build +# SCCACHE_BUCKET: "ossci-compiler-cache-circleci-v2" +# AWS_ACCESS_KEY_ID: $(SCCACHE_ACCESS_KEY_ID) +# AWS_SECRET_ACCESS_KEY: $(SCCACHE_SECRET) +# displayName: 'Build' +# - bash: | +# set +x +# set -e +# export PATH="/home/azureuser_azpcontainer/.local/bin:$PATH" +# .jenkins/pytorch/test.sh +# env: +# BUILD_ENVIRONMENT: pytorch-linux-trusty-py3.5-test +# displayName: 'Test' From 8c6ea6ce68cf1ddf9f9c246950c00c0e7d8dc6af Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Fri, 10 May 2019 18:04:03 -0700 Subject: [PATCH 03/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index 7f0a589a8072..5a069c78c428 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -22,8 +22,8 @@ jobs: submodules: "true" - bash: | set +x - sudo chown -R azureuser_azpcontainer:azureuser_azpcontainer /usr/local - sudo chown -R azureuser_azpcontainer:azureuser_azpcontainer /opt/python + sudo chown -R suo_azpcontainer:suo_azpcontainer /usr/local + sudo chown -R suo_azpcontainer:suo_azpcontainer /opt/python .jenkins/pytorch/build.sh env: BUILD_ENVIRONMENT: pytorch-linux-xenial-cuda9-cudnn7-py3-build @@ -34,7 +34,7 @@ jobs: - bash: | set +x set -e - export PATH="/home/azureuser_azpcontainer/.local/bin:$PATH" + export PATH="/home/suo_azpcontainer/.local/bin:$PATH" .jenkins/pytorch/test.sh env: BUILD_ENVIRONMENT: pytorch-linux-trusty-py3.5-test From 09696a65679390397ec149984781023e95a3103c Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Fri, 10 May 2019 18:04:48 -0700 Subject: [PATCH 04/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 1 + 1 file changed, 1 insertion(+) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index 5a069c78c428..d1926efdef1d 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -24,6 +24,7 @@ jobs: set +x sudo chown -R suo_azpcontainer:suo_azpcontainer /usr/local sudo chown -R suo_azpcontainer:suo_azpcontainer /opt/python + sudo chown -R suo_azpcontainer:suo_azpcontainer /opt/conda .jenkins/pytorch/build.sh env: BUILD_ENVIRONMENT: pytorch-linux-xenial-cuda9-cudnn7-py3-build From 6e4135f8634c7db013de2e73f942ad2eee80a6a9 Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Fri, 10 May 2019 23:46:16 -0700 Subject: [PATCH 05/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index d1926efdef1d..27c4e9efcb7e 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -16,7 +16,7 @@ jobs: image: 308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/pytorch-linux-xenial-cuda9-cudnn7-py3:300 endpoint: aws_docker options: -u 0 - timeoutInMinutes: 120 + timeoutInMinutes: 360 steps: - checkout: self submodules: "true" From 954586daf2c0bd59c3899a8cb7b7ce7287197aa1 Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Fri, 10 May 2019 23:46:26 -0700 Subject: [PATCH 06/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index 27c4e9efcb7e..61b9a0713388 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -38,7 +38,7 @@ jobs: export PATH="/home/suo_azpcontainer/.local/bin:$PATH" .jenkins/pytorch/test.sh env: - BUILD_ENVIRONMENT: pytorch-linux-trusty-py3.5-test + BUILD_ENVIRONMENT: pytorch-linux-xenial-cuda9-cudnn7-py3-test displayName: 'Test' # jobs: From cb7eaa43aa49c414e756abe465e76eadcc79cfaa Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Sat, 11 May 2019 01:31:43 -0700 Subject: [PATCH 07/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 1 + 1 file changed, 1 insertion(+) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index 61b9a0713388..cae7e4dd6630 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -39,6 +39,7 @@ jobs: .jenkins/pytorch/test.sh env: BUILD_ENVIRONMENT: pytorch-linux-xenial-cuda9-cudnn7-py3-test + USE_CUDA_DOCKER_RUNTIME: 1 displayName: 'Test' # jobs: From 02228a3b007cdfe59a7b4d64afc821149d99c87e Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Mon, 13 May 2019 10:51:17 -0700 Subject: [PATCH 08/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index cae7e4dd6630..1a5fb0f1b06a 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -15,7 +15,7 @@ jobs: container: image: 308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/pytorch-linux-xenial-cuda9-cudnn7-py3:300 endpoint: aws_docker - options: -u 0 + options: -u 0 --runtime=nvidia timeoutInMinutes: 360 steps: - checkout: self From ad57c892a78a022e553f06757ad5394534f4a057 Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Mon, 13 May 2019 13:14:59 -0700 Subject: [PATCH 09/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index 1a5fb0f1b06a..ccdc90ba5e3d 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -35,6 +35,34 @@ jobs: - bash: | set +x set -e + # Set up NVIDIA docker repo + curl -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add - + echo "deb https://nvidia.github.io/libnvidia-container/ubuntu16.04/amd64 /" | sudo tee -a /etc/apt/sources.list.d/nvidia-docker.list + echo "deb https://nvidia.github.io/nvidia-container-runtime/ubuntu16.04/amd64 /" | sudo tee -a /etc/apt/sources.list.d/nvidia-docker.list + echo "deb https://nvidia.github.io/nvidia-docker/ubuntu16.04/amd64 /" | sudo tee -a /etc/apt/sources.list.d/nvidia-docker.list + + sudo apt-get -y update + sudo apt-get -y remove linux-image-generic linux-headers-generic linux-generic docker-ce + # WARNING: Docker version is hardcoded here; you must update the + # version number below for docker-ce and nvidia-docker2 to get newer + # versions of Docker. We hardcode these numbers because we kept + # getting broken CI when Docker would update their docker version, + # and nvidia-docker2 would be out of date for a day until they + # released a newer version of their package. + # + # How to figure out what the correct versions of these packages are? + # My preferred method is to start a Docker instance of the correct + # Ubuntu version (e.g., docker run -it ubuntu:16.04) and then ask + # apt what the packages you need are. Note that the CircleCI image + # comes with Docker. + sudo apt-get -y install \ + linux-headers-$(uname -r) \ + linux-image-generic \ + moreutils \ + docker-ce=5:18.09.4~3-0~ubuntu-xenial \ + nvidia-container-runtime=2.0.0+docker18.09.4-1 \ + nvidia-docker2=2.0.3+docker18.09.4-1 \ + expect-dev export PATH="/home/suo_azpcontainer/.local/bin:$PATH" .jenkins/pytorch/test.sh env: From f7580df6b313c838e82c397da0d5766363dad90e Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Mon, 13 May 2019 15:20:20 -0700 Subject: [PATCH 10/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 28 ---------------------------- 1 file changed, 28 deletions(-) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index ccdc90ba5e3d..1a5fb0f1b06a 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -35,34 +35,6 @@ jobs: - bash: | set +x set -e - # Set up NVIDIA docker repo - curl -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add - - echo "deb https://nvidia.github.io/libnvidia-container/ubuntu16.04/amd64 /" | sudo tee -a /etc/apt/sources.list.d/nvidia-docker.list - echo "deb https://nvidia.github.io/nvidia-container-runtime/ubuntu16.04/amd64 /" | sudo tee -a /etc/apt/sources.list.d/nvidia-docker.list - echo "deb https://nvidia.github.io/nvidia-docker/ubuntu16.04/amd64 /" | sudo tee -a /etc/apt/sources.list.d/nvidia-docker.list - - sudo apt-get -y update - sudo apt-get -y remove linux-image-generic linux-headers-generic linux-generic docker-ce - # WARNING: Docker version is hardcoded here; you must update the - # version number below for docker-ce and nvidia-docker2 to get newer - # versions of Docker. We hardcode these numbers because we kept - # getting broken CI when Docker would update their docker version, - # and nvidia-docker2 would be out of date for a day until they - # released a newer version of their package. - # - # How to figure out what the correct versions of these packages are? - # My preferred method is to start a Docker instance of the correct - # Ubuntu version (e.g., docker run -it ubuntu:16.04) and then ask - # apt what the packages you need are. Note that the CircleCI image - # comes with Docker. - sudo apt-get -y install \ - linux-headers-$(uname -r) \ - linux-image-generic \ - moreutils \ - docker-ce=5:18.09.4~3-0~ubuntu-xenial \ - nvidia-container-runtime=2.0.0+docker18.09.4-1 \ - nvidia-docker2=2.0.3+docker18.09.4-1 \ - expect-dev export PATH="/home/suo_azpcontainer/.local/bin:$PATH" .jenkins/pytorch/test.sh env: From e0978760a467fefbf1560f8c8a68c99c8d8cdd66 Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Mon, 13 May 2019 15:21:08 -0700 Subject: [PATCH 11/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index 1a5fb0f1b06a..fd89a0bd225e 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -22,10 +22,7 @@ jobs: submodules: "true" - bash: | set +x - sudo chown -R suo_azpcontainer:suo_azpcontainer /usr/local - sudo chown -R suo_azpcontainer:suo_azpcontainer /opt/python - sudo chown -R suo_azpcontainer:suo_azpcontainer /opt/conda - .jenkins/pytorch/build.sh + sudo .jenkins/pytorch/build.sh env: BUILD_ENVIRONMENT: pytorch-linux-xenial-cuda9-cudnn7-py3-build SCCACHE_BUCKET: "ossci-compiler-cache-circleci-v2" From 440b3314d83a507841af5c1626e4132c2de111aa Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Mon, 13 May 2019 15:24:43 -0700 Subject: [PATCH 12/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index fd89a0bd225e..00c1bf651f67 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -21,7 +21,7 @@ jobs: - checkout: self submodules: "true" - bash: | - set +x + set -ex sudo .jenkins/pytorch/build.sh env: BUILD_ENVIRONMENT: pytorch-linux-xenial-cuda9-cudnn7-py3-build From 0685c1932d537bacf2188c193ce44f16314189bc Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Mon, 13 May 2019 15:26:41 -0700 Subject: [PATCH 13/26] foo --- .jenkins/pytorch/build.sh | 1 + 1 file changed, 1 insertion(+) diff --git a/.jenkins/pytorch/build.sh b/.jenkins/pytorch/build.sh index fbef26a95714..909f76182af5 100755 --- a/.jenkins/pytorch/build.sh +++ b/.jenkins/pytorch/build.sh @@ -6,6 +6,7 @@ # shellcheck disable=SC2034 COMPACT_JOB_NAME="${BUILD_ENVIRONMENT}" +echo "Compact job name: $COMPACT_JOB_NAME" source "$(dirname "${BASH_SOURCE[0]}")/common.sh" From 33703d04bbdf52339b95a9e71414832a6f9b8934 Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Mon, 13 May 2019 15:28:07 -0700 Subject: [PATCH 14/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 1 + 1 file changed, 1 insertion(+) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index 00c1bf651f67..77424d9ca7cc 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -22,6 +22,7 @@ jobs: submodules: "true" - bash: | set -ex + echo $BUILD_ENVIRONMENT sudo .jenkins/pytorch/build.sh env: BUILD_ENVIRONMENT: pytorch-linux-xenial-cuda9-cudnn7-py3-build From c40a88a4c46de0b413d61dd9389fa9ab649a968c Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Mon, 13 May 2019 15:35:25 -0700 Subject: [PATCH 15/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index 77424d9ca7cc..206f1c727d91 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -23,7 +23,10 @@ jobs: - bash: | set -ex echo $BUILD_ENVIRONMENT - sudo .jenkins/pytorch/build.sh + # we should probably remove the sudo -E; this is because the container user + # azure creates is not the same as the docker container owner, and chown -R'ing + # everything takes a long time. + sudo -E .jenkins/pytorch/build.sh env: BUILD_ENVIRONMENT: pytorch-linux-xenial-cuda9-cudnn7-py3-build SCCACHE_BUCKET: "ossci-compiler-cache-circleci-v2" From 869b61248f5928ef1af838447e9438e77d44c4c2 Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Mon, 13 May 2019 15:55:47 -0700 Subject: [PATCH 16/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index 206f1c727d91..88d285e74f77 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -22,11 +22,10 @@ jobs: submodules: "true" - bash: | set -ex - echo $BUILD_ENVIRONMENT # we should probably remove the sudo -E; this is because the container user # azure creates is not the same as the docker container owner, and chown -R'ing # everything takes a long time. - sudo -E .jenkins/pytorch/build.sh + .jenkins/pytorch/build.sh env: BUILD_ENVIRONMENT: pytorch-linux-xenial-cuda9-cudnn7-py3-build SCCACHE_BUCKET: "ossci-compiler-cache-circleci-v2" From 8540e1537a341827ed63e56e321aa9e35b1418ec Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Mon, 13 May 2019 15:58:54 -0700 Subject: [PATCH 17/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index 88d285e74f77..3d383812d1c5 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -21,7 +21,8 @@ jobs: - checkout: self submodules: "true" - bash: | - set -ex + set -x + sudo chown -R /opt/conda # we should probably remove the sudo -E; this is because the container user # azure creates is not the same as the docker container owner, and chown -R'ing # everything takes a long time. From c8f15f2317ec2b0788aa41d7a23f24658dc71d95 Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Mon, 13 May 2019 16:32:34 -0700 Subject: [PATCH 18/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index 3d383812d1c5..913e36eb817e 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -13,9 +13,11 @@ jobs: demands: "gpu" displayName: "Build and Test" container: - image: 308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/pytorch-linux-xenial-cuda9-cudnn7-py3:300 - endpoint: aws_docker - options: -u 0 --runtime=nvidia + image: http://pytorch.azurecr.io/tmp-lpmzxocpdogrtxhbleawtkoqukyezwwv:latest + endpoint: pytorch + # image: 308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/pytorch-linux-xenial-cuda9-cudnn7-py3:300 + # endpoint: aws_docker + # options: -u 0 --runtime=nvidia timeoutInMinutes: 360 steps: - checkout: self From ba4b0f5485c9a1194348bfa670fff25a0f8d386c Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Mon, 13 May 2019 16:38:16 -0700 Subject: [PATCH 19/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index 913e36eb817e..15cf442880c0 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -13,7 +13,7 @@ jobs: demands: "gpu" displayName: "Build and Test" container: - image: http://pytorch.azurecr.io/tmp-lpmzxocpdogrtxhbleawtkoqukyezwwv:latest + image: pytorch.azurecr.io/tmp-lpmzxocpdogrtxhbleawtkoqukyezwwv:latest endpoint: pytorch # image: 308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/pytorch-linux-xenial-cuda9-cudnn7-py3:300 # endpoint: aws_docker From 78a3cf5eb0b3b4abe5fb0c8ce7c613d7124a8874 Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Mon, 13 May 2019 16:52:59 -0700 Subject: [PATCH 20/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 1 + 1 file changed, 1 insertion(+) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index 15cf442880c0..d5123d2928a5 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -15,6 +15,7 @@ jobs: container: image: pytorch.azurecr.io/tmp-lpmzxocpdogrtxhbleawtkoqukyezwwv:latest endpoint: pytorch + options: -u 0 # image: 308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/pytorch-linux-xenial-cuda9-cudnn7-py3:300 # endpoint: aws_docker # options: -u 0 --runtime=nvidia From b1f90a4eb9997ea35e61ec45a54e27e6dfd76a17 Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Mon, 13 May 2019 16:53:08 -0700 Subject: [PATCH 21/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index d5123d2928a5..ea896e404252 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -15,7 +15,7 @@ jobs: container: image: pytorch.azurecr.io/tmp-lpmzxocpdogrtxhbleawtkoqukyezwwv:latest endpoint: pytorch - options: -u 0 + options: -u 0 --runtime=nvidia # image: 308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/pytorch-linux-xenial-cuda9-cudnn7-py3:300 # endpoint: aws_docker # options: -u 0 --runtime=nvidia From 8cd6c88ce29158c1293e947e068c6ff04c8e245e Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Mon, 13 May 2019 16:55:43 -0700 Subject: [PATCH 22/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index ea896e404252..0e41adab3148 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -25,7 +25,8 @@ jobs: submodules: "true" - bash: | set -x - sudo chown -R /opt/conda + id = $(id -u) + sudo chown -R $id:$id /opt/conda # we should probably remove the sudo -E; this is because the container user # azure creates is not the same as the docker container owner, and chown -R'ing # everything takes a long time. From 306a9ae6ddea4fd772941daf81b61adfe66d5686 Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Mon, 13 May 2019 16:58:18 -0700 Subject: [PATCH 23/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index 0e41adab3148..bc1c03289c50 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -25,7 +25,7 @@ jobs: submodules: "true" - bash: | set -x - id = $(id -u) + id=$(id -u) sudo chown -R $id:$id /opt/conda # we should probably remove the sudo -E; this is because the container user # azure creates is not the same as the docker container owner, and chown -R'ing From 3fdec3359dd787e40c9ebf68647c891315824f04 Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Tue, 14 May 2019 09:01:36 -0700 Subject: [PATCH 24/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 1 + 1 file changed, 1 insertion(+) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index bc1c03289c50..3e9ef75fc51c 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -36,6 +36,7 @@ jobs: SCCACHE_BUCKET: "ossci-compiler-cache-circleci-v2" AWS_ACCESS_KEY_ID: $(SCCACHE_ACCESS_KEY_ID) AWS_SECRET_ACCESS_KEY: $(SCCACHE_SECRET) + TORCH_CUDA_ARCH_LIST: 5.2 displayName: 'Build' - bash: | set +x From f48d8e74b8fcbe08e29c452efe42f40218a38402 Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Tue, 14 May 2019 09:43:44 -0700 Subject: [PATCH 25/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 2 ++ 1 file changed, 2 insertions(+) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index 3e9ef75fc51c..170df0eb8c6e 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -16,6 +16,8 @@ jobs: image: pytorch.azurecr.io/tmp-lpmzxocpdogrtxhbleawtkoqukyezwwv:latest endpoint: pytorch options: -u 0 --runtime=nvidia + workspace: + clean: outputs # image: 308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/pytorch-linux-xenial-cuda9-cudnn7-py3:300 # endpoint: aws_docker # options: -u 0 --runtime=nvidia From 0442053f71321297c0f9b63b838862a16a8e95e1 Mon Sep 17 00:00:00 2001 From: Michael Suo Date: Tue, 14 May 2019 10:07:03 -0700 Subject: [PATCH 26/26] Update azure-pipelines.yml for Azure Pipelines --- azure-pipelines.yml | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/azure-pipelines.yml b/azure-pipelines.yml index 170df0eb8c6e..9caaffa3af66 100644 --- a/azure-pipelines.yml +++ b/azure-pipelines.yml @@ -16,8 +16,6 @@ jobs: image: pytorch.azurecr.io/tmp-lpmzxocpdogrtxhbleawtkoqukyezwwv:latest endpoint: pytorch options: -u 0 --runtime=nvidia - workspace: - clean: outputs # image: 308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/pytorch-linux-xenial-cuda9-cudnn7-py3:300 # endpoint: aws_docker # options: -u 0 --runtime=nvidia @@ -25,6 +23,7 @@ jobs: steps: - checkout: self submodules: "true" + clean: "all" - bash: | set -x id=$(id -u)