Properly cleanup processes and queues for MPRS and Fix pause for prefetch (#1075)#1096
Closed
ejguan wants to merge 1 commit intometa-pytorch:mainfrom
Closed
Properly cleanup processes and queues for MPRS and Fix pause for prefetch (#1075)#1096ejguan wants to merge 1 commit intometa-pytorch:mainfrom
pause for prefetch (#1075)#1096ejguan wants to merge 1 commit intometa-pytorch:mainfrom
Conversation
…efetch (meta-pytorch#1075) Summary: Fixes issue about `MPRS.finalize` when `dataloader2.shutdown()` is called ### Changes - DataLoader2 should always clean up `datapipe_iter` at shutdown - Guard `MPRS` to finalize once - Fix the problem of `ConnectionError` when DataLoader early exits - This is caused by `queue` is joined when main/worker/dispatching process exits. No more request/response can be passed across processes. - Consumer process shouldn't join the `req_queue` at exit to make sure producer process can still access the remaining request. And, consumer will close `req_queue` after clean up to prevent any further request sent to queue. - Produce process shouldn't join the `res_queue` at exit to make sure consumer process can still access response. And, producer will close `res_queue` after clean up to prevent any further response sent to queue. - Main (Consumer) <-> Worker (Producer) - Worker (Consumer) -> Dispatching (Producer) - Fix `pause` API for DataLoader2 - Invoke `pause` lazily until the `limit+1` iteration is reached to align with python's iterator behavior. - Make `prefetch.pause` blocking unless there might be potential racing issue. Main thread is paused but prefetch worker is still trying to fetch data from `iter`. - Add tests to validate Pull Request resolved: meta-pytorch#1075 Differential Revision: D44168655 Pulled By: ejguan fbshipit-source-id: 4fa8c34438414cad7f46527355eb0d51866e6b15
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D44168655 |
Contributor
Author
|
This is the reland of #1075. To fix the internal test failures, I have to remove the optional |
NivekT
approved these changes
Mar 17, 2023
Contributor
NivekT
pushed a commit
that referenced
this pull request
Apr 19, 2023
…efetch (#1096) Summary: Pull Request resolved: #1096 Fixes issue about `MPRS.finalize` when `dataloader2.shutdown()` is called ### Changes - DataLoader2 should always clean up `datapipe_iter` at shutdown - Guard `MPRS` to finalize once - Fix the problem of `ConnectionError` when DataLoader early exits - This is caused by `queue` is joined when main/worker/dispatching process exits. No more request/response can be passed across processes. - Consumer process shouldn't join the `req_queue` at exit to make sure producer process can still access the remaining request. And, consumer will close `req_queue` after clean up to prevent any further request sent to queue. - Produce process shouldn't join the `res_queue` at exit to make sure consumer process can still access response. And, producer will close `res_queue` after clean up to prevent any further response sent to queue. - Main (Consumer) <-> Worker (Producer) - Worker (Consumer) -> Dispatching (Producer) - Fix `pause` API for DataLoader2 - Invoke `pause` lazily until the `limit+1` iteration is reached to align with python's iterator behavior. - Make `prefetch.pause` blocking unless there might be potential racing issue. Main thread is paused but prefetch worker is still trying to fetch data from `iter`. - Add tests to validate Pull Request resolved: #1075 Reviewed By: NivekT Differential Revision: D44168655 Pulled By: ejguan fbshipit-source-id: fdfee5c27b512b5c0d5308e53a81b1cb2db70a43
ejguan
added a commit
that referenced
this pull request
Apr 20, 2023
…efetch (#1096) Summary: Pull Request resolved: #1096 Fixes issue about `MPRS.finalize` when `dataloader2.shutdown()` is called ### Changes - DataLoader2 should always clean up `datapipe_iter` at shutdown - Guard `MPRS` to finalize once - Fix the problem of `ConnectionError` when DataLoader early exits - This is caused by `queue` is joined when main/worker/dispatching process exits. No more request/response can be passed across processes. - Consumer process shouldn't join the `req_queue` at exit to make sure producer process can still access the remaining request. And, consumer will close `req_queue` after clean up to prevent any further request sent to queue. - Produce process shouldn't join the `res_queue` at exit to make sure consumer process can still access response. And, producer will close `res_queue` after clean up to prevent any further response sent to queue. - Main (Consumer) <-> Worker (Producer) - Worker (Consumer) -> Dispatching (Producer) - Fix `pause` API for DataLoader2 - Invoke `pause` lazily until the `limit+1` iteration is reached to align with python's iterator behavior. - Make `prefetch.pause` blocking unless there might be potential racing issue. Main thread is paused but prefetch worker is still trying to fetch data from `iter`. - Add tests to validate Pull Request resolved: #1075 Reviewed By: NivekT Differential Revision: D44168655 Pulled By: ejguan fbshipit-source-id: fdfee5c27b512b5c0d5308e53a81b1cb2db70a43
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary:
Fixes issue about
MPRS.finalizewhendataloader2.shutdown()is calledChanges
datapipe_iterat shutdownMPRSto finalize onceConnectionErrorwhen DataLoader early exitsqueueis joined when main/worker/dispatching process exits. No more request/response can be passed across processes.req_queueat exit to make sure producer process can still access the remaining request. And, consumer will closereq_queueafter clean up to prevent any further request sent to queue.res_queueat exit to make sure consumer process can still access response. And, producer will closeres_queueafter clean up to prevent any further response sent to queue.pauseAPI for DataLoader2Invoke(Breaking internal tests)pauselazily until thelimit+1iteration is reached to align with python's iterator behavior.prefetch.pauseblocking unless there might be potential racing issue. Main thread is paused but prefetch worker is still trying to fetch data fromiter.Pull Request resolved: #1075
Differential Revision: D44168655
Pulled By: ejguan