[quant][fx] Cleanup unused to_fp16 check code in lowering #74969

jerryzh168 · 2022-03-30T18:45:38Z

Stack from ghstack (oldest at bottom):

-> [quant][fx] Cleanup unused to_fp16 check code in lowering #74969

Summary:
We can remove the check for fp16 ops now since we confirmed that fp16 ops are not
used

Test Plan:
python test/test_quantization.py TestQuantizeFx
python test/test_quantization.py TestQuantizeFxOps

Reviewers:

Subscribers:

Tasks:

Tags:

Differential Revision: D35258695

Summary: We can remove the check for fp16 ops now since we confirmed that fp16 ops are not used Test Plan: python test/test_quantization.py TestQuantizeFx python test/test_quantization.py TestQuantizeFxOps Reviewers: Subscribers: Tasks: Tags: [ghstack-poisoned]

facebook-github-bot · 2022-03-30T18:45:44Z

🔗 Helpful links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/74969
↩️ [fb-only] Re-run with SSH instructions
Need help or want to give feedback on the CI? Visit our office hours

💊 CI failures summary and remediations

As of commit 2a8b798 (more details on the Dr. CI page):

3/3 failures introduced in this PR

🕵️ 3 new failures recognized by patterns

The following CI failures do not appear to be due to upstream breakages:

pull / deploy-linux-xenial-cuda11.3-py3.7-gcc7 / build (1/3)

Step: "Build" (full log | diagnosis details | 🔁 rerun)

2022-03-30T19:03:42.1775084Z �[36;1m echo "ERR...t available for the merge-base of your branch"�[0m

2022-03-30T19:03:42.1772238Z �[36;1mfi�[0m
2022-03-30T19:03:42.1772464Z �[36;1m# Covers the case where a previous tag doesn't exist for the tree�[0m
2022-03-30T19:03:42.1772803Z �[36;1m# this is only really applicable on trees that don't have `.circleci/docker` at its merge base, i.e. nightly�[0m
2022-03-30T19:03:42.1773113Z �[36;1mif ! git rev-parse "$MERGE_BASE:.circleci/docker"; then�[0m
2022-03-30T19:03:42.1773455Z �[36;1m  echo "Directory '.circleci/docker' not found in commit $MERGE_BASE, you should probably rebase onto a more recent commit"�[0m
2022-03-30T19:03:42.1773735Z �[36;1m  exit 1�[0m
2022-03-30T19:03:42.1773898Z �[36;1mfi�[0m
2022-03-30T19:03:42.1774119Z �[36;1mPREVIOUS_DOCKER_TAG=$(git rev-parse "$MERGE_BASE:.circleci/docker")�[0m
2022-03-30T19:03:42.1774451Z �[36;1m# If no image exists but the hash is the same as the previous hash then we should error out here�[0m
2022-03-30T19:03:42.1774755Z �[36;1mif [[ "${PREVIOUS_DOCKER_TAG}" = "${DOCKER_TAG}" ]]; then�[0m
2022-03-30T19:03:42.1775084Z �[36;1m  echo "ERROR: Something has gone wrong and the previous image isn't available for the merge-base of your branch"�[0m
2022-03-30T19:03:42.1775415Z �[36;1m  echo "       contact the PyTorch team to restore the original images"�[0m
2022-03-30T19:03:42.1775730Z �[36;1m  exit 1�[0m
2022-03-30T19:03:42.1775897Z �[36;1mfi�[0m
2022-03-30T19:03:42.1776084Z �[36;1mecho ::set-output name=rebuild::yes�[0m
2022-03-30T19:03:42.1787048Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0}
2022-03-30T19:03:42.1787269Z env:
2022-03-30T19:03:42.1787412Z   IN_CI: 1
2022-03-30T19:03:42.1787574Z   IS_GHA: 1
2022-03-30T19:03:42.1787790Z   BASE_REVISION: 79c00de1bc080732ca752d166a30f546104cb7ce
2022-03-30T19:03:42.1788193Z   DOCKER_IMAGE: 308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/pytorch-linux-xenial-cuda11.3-cudnn8-py3-gcc7:a2c09c6009bb8a10cbb45a8c5f7c573593289be0

pull / linux-xenial-cuda11.3-py3.7-gcc7-bazel-test / build-and-test (2/3)

Step: "Build" (full log | diagnosis details | 🔁 rerun)

2022-03-30T19:03:45.6025855Z �[36;1m echo "ERR...t available for the merge-base of your branch"�[0m

2022-03-30T19:03:45.6022599Z �[36;1mfi�[0m
2022-03-30T19:03:45.6022842Z �[36;1m# Covers the case where a previous tag doesn't exist for the tree�[0m
2022-03-30T19:03:45.6023199Z �[36;1m# this is only really applicable on trees that don't have `.circleci/docker` at its merge base, i.e. nightly�[0m
2022-03-30T19:03:45.6023526Z �[36;1mif ! git rev-parse "$MERGE_BASE:.circleci/docker"; then�[0m
2022-03-30T19:03:45.6023887Z �[36;1m  echo "Directory '.circleci/docker' not found in commit $MERGE_BASE, you should probably rebase onto a more recent commit"�[0m
2022-03-30T19:03:45.6024356Z �[36;1m  exit 1�[0m
2022-03-30T19:03:45.6024589Z �[36;1mfi�[0m
2022-03-30T19:03:45.6024831Z �[36;1mPREVIOUS_DOCKER_TAG=$(git rev-parse "$MERGE_BASE:.circleci/docker")�[0m
2022-03-30T19:03:45.6025178Z �[36;1m# If no image exists but the hash is the same as the previous hash then we should error out here�[0m
2022-03-30T19:03:45.6025508Z �[36;1mif [[ "${PREVIOUS_DOCKER_TAG}" = "${DOCKER_TAG}" ]]; then�[0m
2022-03-30T19:03:45.6025855Z �[36;1m  echo "ERROR: Something has gone wrong and the previous image isn't available for the merge-base of your branch"�[0m
2022-03-30T19:03:45.6026285Z �[36;1m  echo "       contact the PyTorch team to restore the original images"�[0m
2022-03-30T19:03:45.6026531Z �[36;1m  exit 1�[0m
2022-03-30T19:03:45.6026703Z �[36;1mfi�[0m
2022-03-30T19:03:45.6026901Z �[36;1mecho ::set-output name=rebuild::yes�[0m
2022-03-30T19:03:45.6037128Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0}
2022-03-30T19:03:45.6037359Z env:
2022-03-30T19:03:45.6037513Z   IN_CI: 1
2022-03-30T19:03:45.6037683Z   IS_GHA: 1
2022-03-30T19:03:45.6037875Z   GIT_DEFAULT_BRANCH: master
2022-03-30T19:03:45.6038182Z   BASE_REVISION: 79c00de1bc080732ca752d166a30f546104cb7ce

pull / linux-xenial-cuda11.3-py3.7-gcc7 / build (3/3)

Step: "Build" (full log | diagnosis details | 🔁 rerun)

2022-03-30T19:03:51.3041854Z �[36;1m echo "ERR...t available for the merge-base of your branch"�[0m

2022-03-30T19:03:51.3038879Z �[36;1mfi�[0m
2022-03-30T19:03:51.3039112Z �[36;1m# Covers the case where a previous tag doesn't exist for the tree�[0m
2022-03-30T19:03:51.3039466Z �[36;1m# this is only really applicable on trees that don't have `.circleci/docker` at its merge base, i.e. nightly�[0m
2022-03-30T19:03:51.3039794Z �[36;1mif ! git rev-parse "$MERGE_BASE:.circleci/docker"; then�[0m
2022-03-30T19:03:51.3040153Z �[36;1m  echo "Directory '.circleci/docker' not found in commit $MERGE_BASE, you should probably rebase onto a more recent commit"�[0m
2022-03-30T19:03:51.3040444Z �[36;1m  exit 1�[0m
2022-03-30T19:03:51.3040615Z �[36;1mfi�[0m
2022-03-30T19:03:51.3040845Z �[36;1mPREVIOUS_DOCKER_TAG=$(git rev-parse "$MERGE_BASE:.circleci/docker")�[0m
2022-03-30T19:03:51.3041192Z �[36;1m# If no image exists but the hash is the same as the previous hash then we should error out here�[0m
2022-03-30T19:03:51.3041509Z �[36;1mif [[ "${PREVIOUS_DOCKER_TAG}" = "${DOCKER_TAG}" ]]; then�[0m
2022-03-30T19:03:51.3041854Z �[36;1m  echo "ERROR: Something has gone wrong and the previous image isn't available for the merge-base of your branch"�[0m
2022-03-30T19:03:51.3042201Z �[36;1m  echo "       contact the PyTorch team to restore the original images"�[0m
2022-03-30T19:03:51.3042736Z �[36;1m  exit 1�[0m
2022-03-30T19:03:51.3042910Z �[36;1mfi�[0m
2022-03-30T19:03:51.3043104Z �[36;1mecho ::set-output name=rebuild::yes�[0m
2022-03-30T19:03:51.3053719Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0}
2022-03-30T19:03:51.3053947Z env:
2022-03-30T19:03:51.3054098Z   IN_CI: 1
2022-03-30T19:03:51.3054266Z   IS_GHA: 1
2022-03-30T19:03:51.3054571Z   BASE_REVISION: 79c00de1bc080732ca752d166a30f546104cb7ce
2022-03-30T19:03:51.3054995Z   DOCKER_IMAGE: 308535385114.dkr.ecr.us-east-1.amazonaws.com/pytorch/pytorch-linux-xenial-cuda11.3-cudnn8-py3-gcc7:a2c09c6009bb8a10cbb45a8c5f7c573593289be0

This comment was automatically generated by Dr. CI (expand for details).

Please report bugs/suggestions to the (internal) Dr. CI Users group.

Click here to manually regenerate this comment.

jerryzh168 · 2022-03-30T18:54:37Z

@jerryzh168 has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

Summary: We can remove the check for fp16 ops now since we confirmed that fp16 ops are not used Test Plan: python test/test_quantization.py TestQuantizeFx python test/test_quantization.py TestQuantizeFxOps Reviewers: Subscribers: Tasks: Tags: Differential Revision: [D35258695](https://our.internmc.facebook.com/intern/diff/D35258695) [ghstack-poisoned]

Summary: We can remove the check for fp16 ops now since we confirmed that fp16 ops are not used Test Plan: python test/test_quantization.py TestQuantizeFx python test/test_quantization.py TestQuantizeFxOps Reviewers: Subscribers: Tasks: Tags: ghstack-source-id: 8930486 Pull Request resolved: #74969

jerryzh168 · 2022-03-30T19:02:41Z

@jerryzh168 has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.

Summary: Pull Request resolved: #74969 We can remove the check for fp16 ops now since we confirmed that fp16 ops are not used Test Plan: python test/test_quantization.py TestQuantizeFx python test/test_quantization.py TestQuantizeFxOps Imported from OSS Reviewed By: andrewor14 Differential Revision: D35258695 fbshipit-source-id: 2297696493feb62a4c959e7fbdd6123f59615ef1

malfet · 2022-03-31T10:50:41Z

Reverting as it broke plethora of tests, for example: https://github.com/pytorch/pytorch/runs/5766517514?check_suite_focus=true

Surprisingly, there are is jobs associated with the latest pull request workflow for the commit: https://github.com/pytorch/pytorch/actions/runs/2066837788

facebook-github-bot · 2022-03-31T10:55:34Z

This pull request has been reverted by 9070d363d7532927acd732fe8e02fd5dc340c089. To re-land this change, please open another pull request, assignthe same reviewers, fix the CI failures that caused the revert and make sure that the failing CI runs on the PR by applying the proper ciflow label (e.g., ciflow/trunk).

facebook-github-bot · 2022-03-31T10:56:06Z

This pull request has been reverted by 0b845bb. To re-land this change, please open another pull request, assignthe same reviewers, fix the CI failures that caused the revert and make sure that the failing CI runs on the PR by applying the proper ciflow label (e.g., ciflow/trunk).

jerryzh168 · 2022-04-01T18:55:32Z

I see, it did not show up because there was some problem with the CI...maybe I should test manually first

facebook-github-bot · 2022-04-05T00:55:08Z

This pull request has been reverted by 0b845bb. To re-land this change, please open another pull request, assignthe same reviewers, fix the CI failures that caused the revert and make sure that the failing CI runs on the PR by applying the proper ciflow label (e.g., ciflow/trunk).

facebook-github-bot added the cla signed label Mar 30, 2022

facebook-github-bot added the module: fx label Mar 30, 2022

jerryzh168 requested review from andrewor14 and vkuzo and removed request for andrewor14 March 30, 2022 19:00

andrewor14 approved these changes Mar 30, 2022

View reviewed changes

pytorchmergebot closed this in ec6f767 Mar 31, 2022

facebook-github-bot added the Reverted label Mar 31, 2022

facebook-github-bot deleted the gh/jerryzh168/763/head branch April 3, 2022 14:16

zengk95 mentioned this pull request Apr 5, 2022

[Meta] CI Revert Tracker #66178

Closed

WBobby mentioned this pull request Aug 17, 2022

Add ROCm5.2.3/AMDGPU support for PyTorch WBobby/pytorch#2

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[quant][fx] Cleanup unused to_fp16 check code in lowering #74969

[quant][fx] Cleanup unused to_fp16 check code in lowering #74969

Uh oh!

jerryzh168 commented Mar 30, 2022 •

edited

Loading

Uh oh!

facebook-github-bot commented Mar 30, 2022 •

edited

Loading

Uh oh!

jerryzh168 commented Mar 30, 2022

Uh oh!

jerryzh168 commented Mar 30, 2022

Uh oh!

malfet commented Mar 31, 2022 •

edited

Loading

Uh oh!

facebook-github-bot commented Mar 31, 2022

Uh oh!

facebook-github-bot commented Mar 31, 2022

Uh oh!

jerryzh168 commented Apr 1, 2022

Uh oh!

facebook-github-bot commented Apr 5, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

[quant][fx] Cleanup unused to_fp16 check code in lowering #74969

[quant][fx] Cleanup unused to_fp16 check code in lowering #74969

Uh oh!

Conversation

jerryzh168 commented Mar 30, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

facebook-github-bot commented Mar 30, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful links

💊 CI failures summary and remediations

🕵️ 3 new failures recognized by patterns

pull / deploy-linux-xenial-cuda11.3-py3.7-gcc7 / build (1/3)

pull / linux-xenial-cuda11.3-py3.7-gcc7-bazel-test / build-and-test (2/3)

pull / linux-xenial-cuda11.3-py3.7-gcc7 / build (3/3)

Uh oh!

jerryzh168 commented Mar 30, 2022

Uh oh!

jerryzh168 commented Mar 30, 2022

Uh oh!

malfet commented Mar 31, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

facebook-github-bot commented Mar 31, 2022

Uh oh!

facebook-github-bot commented Mar 31, 2022

Uh oh!

jerryzh168 commented Apr 1, 2022

Uh oh!

facebook-github-bot commented Apr 5, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

jerryzh168 commented Mar 30, 2022 •

edited

Loading

facebook-github-bot commented Mar 30, 2022 •

edited

Loading

malfet commented Mar 31, 2022 •

edited

Loading