[xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks#39428
Closed
ezhulenev wants to merge 2 commits intoopenxla:mainfrom
Closed
[xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks#39428ezhulenev wants to merge 2 commits intoopenxla:mainfrom
ezhulenev wants to merge 2 commits intoopenxla:mainfrom
Conversation
penpornk
approved these changes
Mar 18, 2026
Member
penpornk
left a comment
penpornk
left a comment
There was a problem hiding this comment.
Approving to test internally.
copybara-service Bot
pushed a commit
to tensorflow/tensorflow
that referenced
this pull request
Mar 18, 2026
…e memcpy thunks Imported from GitHub PR openxla/xla#39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt openxla/xla#39006) Copybara import of the project: -- e98a6f854b2ed999c77737799d8f9de5e637e180 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- bbad4c5de99f45d7bcf8b6b05c09c6ddb53e7184 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#39428 from ezhulenev:async-execution-3 bbad4c5de99f45d7bcf8b6b05c09c6ddb53e7184 PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
that referenced
this pull request
Mar 18, 2026
…e memcpy thunks Imported from GitHub PR #39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt #39006) Copybara import of the project: -- e98a6f8 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- bbad4c5 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=#39428 from ezhulenev:async-execution-3 bbad4c5 PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
that referenced
this pull request
Mar 18, 2026
…e memcpy thunks Imported from GitHub PR #39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt #39006) Copybara import of the project: -- e98a6f8 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- bbad4c5 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=#39428 from ezhulenev:async-execution-3 bbad4c5 PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
to tensorflow/tensorflow
that referenced
this pull request
Mar 18, 2026
…e memcpy thunks Imported from GitHub PR openxla/xla#39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt openxla/xla#39006) Copybara import of the project: -- e98a6f854b2ed999c77737799d8f9de5e637e180 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- bbad4c5de99f45d7bcf8b6b05c09c6ddb53e7184 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#39428 from ezhulenev:async-execution-3 bbad4c5de99f45d7bcf8b6b05c09c6ddb53e7184 PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
that referenced
this pull request
Mar 18, 2026
…e memcpy thunks Imported from GitHub PR #39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt #39006) Copybara import of the project: -- e98a6f8 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- bbad4c5 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=#39428 from ezhulenev:async-execution-3 bbad4c5 PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
to tensorflow/tensorflow
that referenced
this pull request
Mar 18, 2026
…e memcpy thunks Imported from GitHub PR openxla/xla#39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt openxla/xla#39006) Copybara import of the project: -- e98a6f854b2ed999c77737799d8f9de5e637e180 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- bbad4c5de99f45d7bcf8b6b05c09c6ddb53e7184 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#39428 from ezhulenev:async-execution-3 bbad4c5de99f45d7bcf8b6b05c09c6ddb53e7184 PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
that referenced
this pull request
Mar 18, 2026
…e memcpy thunks Imported from GitHub PR #39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt #39006) Copybara import of the project: -- e98a6f8 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- bbad4c5 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=#39428 from ezhulenev:async-execution-3 bbad4c5 PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
that referenced
this pull request
Mar 20, 2026
…e memcpy thunks Imported from GitHub PR #39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt #39006) Copybara import of the project: -- e98a6f8 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- bbad4c5 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=#39428 from ezhulenev:async-execution-3 bbad4c5 PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
to tensorflow/tensorflow
that referenced
this pull request
Mar 20, 2026
…e memcpy thunks Imported from GitHub PR openxla/xla#39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt openxla/xla#39006) Copybara import of the project: -- e98a6f854b2ed999c77737799d8f9de5e637e180 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- bbad4c5de99f45d7bcf8b6b05c09c6ddb53e7184 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#39428 from ezhulenev:async-execution-3 bbad4c5de99f45d7bcf8b6b05c09c6ddb53e7184 PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
that referenced
this pull request
Mar 20, 2026
…e memcpy thunks Imported from GitHub PR #39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt #39006) Copybara import of the project: -- e98a6f8 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- bbad4c5 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=#39428 from ezhulenev:async-execution-3 bbad4c5 PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
that referenced
this pull request
Mar 20, 2026
…e memcpy thunks Imported from GitHub PR #39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt #39006) Copybara import of the project: -- e98a6f8 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- bbad4c5 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=#39428 from ezhulenev:async-execution-3 bbad4c5 PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
to tensorflow/tensorflow
that referenced
this pull request
Mar 20, 2026
…e memcpy thunks Imported from GitHub PR openxla/xla#39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt openxla/xla#39006) Copybara import of the project: -- e98a6f854b2ed999c77737799d8f9de5e637e180 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- bbad4c5de99f45d7bcf8b6b05c09c6ddb53e7184 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#39428 from ezhulenev:async-execution-3 bbad4c5de99f45d7bcf8b6b05c09c6ddb53e7184 PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
that referenced
this pull request
Mar 20, 2026
…e memcpy thunks Imported from GitHub PR #39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt #39006) Copybara import of the project: -- e98a6f8 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- bbad4c5 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=#39428 from ezhulenev:async-execution-3 bbad4c5 PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
to tensorflow/tensorflow
that referenced
this pull request
Mar 20, 2026
…e memcpy thunks Imported from GitHub PR openxla/xla#39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt openxla/xla#39006) Copybara import of the project: -- e98a6f854b2ed999c77737799d8f9de5e637e180 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- bbad4c5de99f45d7bcf8b6b05c09c6ddb53e7184 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#39428 from ezhulenev:async-execution-3 bbad4c5de99f45d7bcf8b6b05c09c6ddb53e7184 PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
that referenced
this pull request
Mar 20, 2026
…e memcpy thunks Imported from GitHub PR #39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt #39006) Copybara import of the project: -- e98a6f8 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- bbad4c5 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=#39428 from ezhulenev:async-execution-3 bbad4c5 PiperOrigin-RevId: 885647980
ezhulenev
force-pushed
the
async-execution-3
branch
from
bbad4c5 to
1dc71b5
Compare
apivovarov
approved these changes
Mar 23, 2026
copybara-service Bot
pushed a commit
that referenced
this pull request
Mar 23, 2026
…e memcpy thunks Imported from GitHub PR #39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt #39006) Copybara import of the project: -- c7b9c25 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- 1dc71b5 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=#39428 from ezhulenev:async-execution-3 1dc71b5 PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
to tensorflow/tensorflow
that referenced
this pull request
Mar 23, 2026
…e memcpy thunks Imported from GitHub PR openxla/xla#39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt openxla/xla#39006) Copybara import of the project: -- c7b9c250652aa597cf822cde8f4b3144ca24d5ce by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- 1dc71b5e3336a58936b35799eeff4f0d85772f9b by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#39428 from ezhulenev:async-execution-3 1dc71b5e3336a58936b35799eeff4f0d85772f9b PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
that referenced
this pull request
Mar 23, 2026
…e memcpy thunks Imported from GitHub PR #39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt #39006) Copybara import of the project: -- c7b9c25 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- 1dc71b5 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=#39428 from ezhulenev:async-execution-3 1dc71b5 PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
to tensorflow/tensorflow
that referenced
this pull request
Mar 23, 2026
…e memcpy thunks Imported from GitHub PR openxla/xla#39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt openxla/xla#39006) Copybara import of the project: -- c7b9c250652aa597cf822cde8f4b3144ca24d5ce by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- 1dc71b5e3336a58936b35799eeff4f0d85772f9b by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#39428 from ezhulenev:async-execution-3 1dc71b5e3336a58936b35799eeff4f0d85772f9b PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
that referenced
this pull request
Mar 23, 2026
…e memcpy thunks Imported from GitHub PR #39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt #39006) Copybara import of the project: -- c7b9c25 by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- 1dc71b5 by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=#39428 from ezhulenev:async-execution-3 1dc71b5 PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
to tensorflow/tensorflow
that referenced
this pull request
Mar 23, 2026
…e memcpy thunks Imported from GitHub PR openxla/xla#39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt openxla/xla#39006) Copybara import of the project: -- c7b9c250652aa597cf822cde8f4b3144ca24d5ce by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- 1dc71b5e3336a58936b35799eeff4f0d85772f9b by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#39428 from ezhulenev:async-execution-3 1dc71b5e3336a58936b35799eeff4f0d85772f9b PiperOrigin-RevId: 885647980
copybara-service Bot
pushed a commit
to tensorflow/tensorflow
that referenced
this pull request
Mar 23, 2026
…e memcpy thunks Imported from GitHub PR openxla/xla#39428 Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk. + add a fix for correct execution stream resolving (previous attempt openxla/xla#39006) Copybara import of the project: -- c7b9c250652aa597cf822cde8f4b3144ca24d5ce by Eugene Zhulenev <ezhulenev@openxla.org>: [xla:gpu] Use generic AsynStart/Done thunks for host/device memcpy thunks -- 1dc71b5e3336a58936b35799eeff4f0d85772f9b by Eugene Zhulenev <ezhulenev@openxla.org>: Use params.stream isntead of resolving it from attributes Merging this change closes #39428 PiperOrigin-RevId: 888338491
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.
Replace copy thunk async events with generic AsyncStartThunk/AsyncDoneThunk for H2D/D2H copies. Remove CopyDoneThunk.