Enable RRef timeout for tensorpipe #39531
Closed
Conversation
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
Enables RRef timeout support in TP agent by having TP agent mark timeout errors with `makeRPCError` API. Also does some refactoring so TP agent can print out the timeout for each future that has timed out. Differential Revision: [D21881475](https://our.internmc.facebook.com/intern/diff/D21881475/) [ghstack-poisoned]
rohan-varma
requested review from
beauby,
jiayisuse,
lw,
mrshenli,
osalpekar,
pritamdamania87 and
zhaojuanmao
as code owners
Closed
rohan-varma
added a commit
that referenced
this pull request
Jun 4, 2020
Enables RRef timeout support in TP agent by having TP agent mark timeout errors with `makeRPCError` API. Also does some refactoring so TP agent can print out the timeout for each future that has timed out. Differential Revision: [D21881475](https://our.internmc.facebook.com/intern/diff/D21881475/) ghstack-source-id: 105273154 Pull Request resolved: #39531
|
@lw Do you mind taking a look at this diff? Thanks! |
osalpekar
reviewed
Jun 5, 2020
| @@ -23,6 +24,8 @@ const std::string kNumIdleThreads = "agent.num_idle_threads"; | |||
| const std::string kClientActiveCalls = "agent.client_active_calls"; | |||
| const std::string kServerActiveCalls = "agent.server_active_calls"; | |||
| const std::string kServerActiveAsyncCalls = "agent.server_active_async_calls"; | |||
| const std::string kTensorPipeRPCTimeoutErrorStr = | |||
There was a problem hiding this comment.
nit: kTensorPipeRpcTimeoutErrorStr
There was a problem hiding this comment.
nit: no need for the TensorPipe prefix, this is in an anonymous namespace, it's local to this file
| @@ -418,7 +421,8 @@ std::shared_ptr<FutureMessage> TensorPipeAgent::send( | |||
| { | |||
| std::unique_lock<std::mutex> lock(timeoutMapMutex_); | |||
| auto& timeoutFuturesVector = timeoutMap_[expirationTime]; | |||
| timeoutFuturesVector.emplace_back(futureResponseMessage); | |||
| timeoutFuturesVector.emplace_back( | |||
| std::make_pair(futureResponseMessage, timeout)); | |||
There was a problem hiding this comment.
I think you might not need to call std::make_pair with emplace_back?
| @@ -2833,7 +2833,6 @@ def test_return_future(self): | |||
| return_future, | |||
| ) | |||
|
|
|||
| @_skip_if_tensorpipe_agent | |||
| @dist_init | |||
There was a problem hiding this comment.
Has this been stress-tested? Just checking to ensure we don't introduce more flaky TensorPipe agent tests
There was a problem hiding this comment.
Ran it with 500 stresstests and they all passed.
Enables RRef timeout support in TP agent by having TP agent mark timeout errors with `makeRPCError` API. Also does some refactoring so TP agent can print out the timeout for each future that has timed out. Differential Revision: [D21881475](https://our.internmc.facebook.com/intern/diff/D21881475/) [ghstack-poisoned]
rohan-varma
added a commit
that referenced
this pull request
Jun 5, 2020
Pull Request resolved: #39531 Enables RRef timeout support in TP agent by having TP agent mark timeout errors with `makeRPCError` API. Also does some refactoring so TP agent can print out the timeout for each future that has timed out. ghstack-source-id: 105377025 Differential Revision: [D21881475](https://our.internmc.facebook.com/intern/diff/D21881475/)
Enables RRef timeout support in TP agent by having TP agent mark timeout errors with `makeRPCError` API. Also does some refactoring so TP agent can print out the timeout for each future that has timed out. Differential Revision: [D21881475](https://our.internmc.facebook.com/intern/diff/D21881475/) [ghstack-poisoned]
💊 CI failures summary and remediationsAs of commit 5661c87 (more details on the Dr. CI page): 💚 💚 Looks good so far! There are no failures yet. 💚 💚 This comment was automatically generated by Dr. CI (expand for details).Follow this link to opt-out of these comments for your Pull Requests.Please report bugs/suggestions on the GitHub issue tracker or post in the (internal) Dr. CI Users group. This comment has been revised 3 times. |
Enables RRef timeout support in TP agent by having TP agent mark timeout errors with `makeRPCError` API. Also does some refactoring so TP agent can print out the timeout for each future that has timed out. Differential Revision: [D21881475](https://our.internmc.facebook.com/intern/diff/D21881475/) [ghstack-poisoned]
rohan-varma
added a commit
that referenced
this pull request
Jun 8, 2020
Pull Request resolved: #39531 Enables RRef timeout support in TP agent by having TP agent mark timeout errors with `makeRPCError` API. Also does some refactoring so TP agent can print out the timeout for each future that has timed out. ghstack-source-id: 105461555 Differential Revision: [D21881475](https://our.internmc.facebook.com/intern/diff/D21881475/)
|
This pull request has been merged in e033db0. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Stack from ghstack:
Enables RRef timeout support in TP agent by having TP agent mark
timeout errors with
makeRPCErrorAPI. Also does some refactoring so TP agentcan print out the timeout for each future that has timed out.
Differential Revision: D21881475