Skip to content

Conversation

@ashiqimranintel
Copy link
Contributor

This PR will enable sycl/oneapi plugin support for Intel GPU

Copy link
Member

@beckerhe beckerhe left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Overall looks good to me. Thank you!

xla::PjRtPlatformId platform_id;
std::string platform_name;

std::string plugin_platform = std::string(PJRT_GPU_PLUGIN_PLATFORM_NAME);
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can you make this an absl::string_view?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

done

#ifdef TENSORFLOW_USE_ROCM
if (allocator_option == "cuda_async") {
VLOG(1) << "cuda_async allocator not available on ROCm!";
VLOG(1) << "cuda_async allocator not available on ROCm! and SYCL!";
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nit: There is an exclamation mark too many.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

done

copybara-service bot pushed a commit that referenced this pull request Jan 7, 2026
…ntel GPU

Imported from GitHub PR #35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a2 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

FUTURE_COPYBARA_INTEGRATE_REVIEW=#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd
PiperOrigin-RevId: 853258690
copybara-service bot pushed a commit to tensorflow/tensorflow that referenced this pull request Jan 7, 2026
…ntel GPU

Imported from GitHub PR openxla/xla#35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a21e9e59c38e961ccd68b036a2c52403728 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1abcd5b84fc503fdc6a2502adeebf813d4 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd19d4e6ba9d521003b4f047e687feffd3d by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd19d4e6ba9d521003b4f047e687feffd3d
PiperOrigin-RevId: 853258690
copybara-service bot pushed a commit that referenced this pull request Jan 7, 2026
…ntel GPU

Imported from GitHub PR #35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a2 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

FUTURE_COPYBARA_INTEGRATE_REVIEW=#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd
PiperOrigin-RevId: 853258690
copybara-service bot pushed a commit to tensorflow/tensorflow that referenced this pull request Jan 7, 2026
…ntel GPU

Imported from GitHub PR openxla/xla#35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a21e9e59c38e961ccd68b036a2c52403728 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1abcd5b84fc503fdc6a2502adeebf813d4 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd19d4e6ba9d521003b4f047e687feffd3d by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd19d4e6ba9d521003b4f047e687feffd3d
PiperOrigin-RevId: 853258690
copybara-service bot pushed a commit that referenced this pull request Jan 7, 2026
…ntel GPU

Imported from GitHub PR #35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a2 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

FUTURE_COPYBARA_INTEGRATE_REVIEW=#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd
PiperOrigin-RevId: 853258690
copybara-service bot pushed a commit to tensorflow/tensorflow that referenced this pull request Jan 7, 2026
…ntel GPU

Imported from GitHub PR openxla/xla#35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a21e9e59c38e961ccd68b036a2c52403728 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1abcd5b84fc503fdc6a2502adeebf813d4 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd19d4e6ba9d521003b4f047e687feffd3d by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd19d4e6ba9d521003b4f047e687feffd3d
PiperOrigin-RevId: 853258690
copybara-service bot pushed a commit that referenced this pull request Jan 7, 2026
…ntel GPU

Imported from GitHub PR #35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a2 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

FUTURE_COPYBARA_INTEGRATE_REVIEW=#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd
PiperOrigin-RevId: 853258690
copybara-service bot pushed a commit to tensorflow/tensorflow that referenced this pull request Jan 7, 2026
…ntel GPU

Imported from GitHub PR openxla/xla#35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a21e9e59c38e961ccd68b036a2c52403728 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1abcd5b84fc503fdc6a2502adeebf813d4 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd19d4e6ba9d521003b4f047e687feffd3d by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

Reverts 989c58d

FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd19d4e6ba9d521003b4f047e687feffd3d
PiperOrigin-RevId: 853258690
copybara-service bot pushed a commit that referenced this pull request Jan 8, 2026
…ntel GPU

Imported from GitHub PR #35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a2 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

FUTURE_COPYBARA_INTEGRATE_REVIEW=#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd
PiperOrigin-RevId: 853719463
copybara-service bot pushed a commit to tensorflow/tensorflow that referenced this pull request Jan 8, 2026
…ntel GPU

Imported from GitHub PR openxla/xla#35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a21e9e59c38e961ccd68b036a2c52403728 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1abcd5b84fc503fdc6a2502adeebf813d4 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd19d4e6ba9d521003b4f047e687feffd3d by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd19d4e6ba9d521003b4f047e687feffd3d
PiperOrigin-RevId: 853719463
@ashiqimranintel
Copy link
Contributor Author

@beckerhe , added a commit that is to fix the JAX failures and also that the bazel query failure is address by the tools PR that has been merged google-ml-infra/rules_ml_toolchain#168

copybara-service bot pushed a commit that referenced this pull request Jan 9, 2026
…ntel GPU

Imported from GitHub PR #35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a2 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

FUTURE_COPYBARA_INTEGRATE_REVIEW=#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd
PiperOrigin-RevId: 853719463
copybara-service bot pushed a commit to tensorflow/tensorflow that referenced this pull request Jan 9, 2026
…ntel GPU

Imported from GitHub PR openxla/xla#35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a21e9e59c38e961ccd68b036a2c52403728 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1abcd5b84fc503fdc6a2502adeebf813d4 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd19d4e6ba9d521003b4f047e687feffd3d by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd19d4e6ba9d521003b4f047e687feffd3d
PiperOrigin-RevId: 853719463
copybara-service bot pushed a commit that referenced this pull request Jan 13, 2026
…ntel GPU

Imported from GitHub PR #35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a2 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

FUTURE_COPYBARA_INTEGRATE_REVIEW=#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd
PiperOrigin-RevId: 853719463
copybara-service bot pushed a commit to tensorflow/tensorflow that referenced this pull request Jan 13, 2026
…ntel GPU

Imported from GitHub PR openxla/xla#35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a21e9e59c38e961ccd68b036a2c52403728 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1abcd5b84fc503fdc6a2502adeebf813d4 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd19d4e6ba9d521003b4f047e687feffd3d by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd19d4e6ba9d521003b4f047e687feffd3d
PiperOrigin-RevId: 853719463
@penpornk
Copy link
Member

@ashiqimranintel Could you please help take a look at the CI failures?

copybara-service bot pushed a commit that referenced this pull request Jan 13, 2026
…ntel GPU

Imported from GitHub PR #35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a2 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

FUTURE_COPYBARA_INTEGRATE_REVIEW=#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd
PiperOrigin-RevId: 853719463
copybara-service bot pushed a commit to tensorflow/tensorflow that referenced this pull request Jan 13, 2026
…ntel GPU

Imported from GitHub PR openxla/xla#35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a21e9e59c38e961ccd68b036a2c52403728 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1abcd5b84fc503fdc6a2502adeebf813d4 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd19d4e6ba9d521003b4f047e687feffd3d by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

Reverts bbea049

FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd19d4e6ba9d521003b4f047e687feffd3d
PiperOrigin-RevId: 853719463
@ashiqimranintel
Copy link
Contributor Author

@penpornk , the failures are unrelated to this PR.

@ashiqimranintel
Copy link
Contributor Author

@penpornk can you check again?

@ashiqimranintel
Copy link
Contributor Author

@penpornk , I resolved the conflict, can you check again?

@penpornk penpornk added the kokoro:force-run Forces CI to rerun label Jan 19, 2026
@penpornk
Copy link
Member

Sorry for the delay. Rerunning the tests now.

@ashiqimranintel
Copy link
Contributor Author

@penpornk , the errors are unrelated to my PR, can you check again?

copybara-service bot pushed a commit that referenced this pull request Jan 26, 2026
…ntel GPU

Imported from GitHub PR #35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a2 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

FUTURE_COPYBARA_INTEGRATE_REVIEW=#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd
PiperOrigin-RevId: 853719463
copybara-service bot pushed a commit to tensorflow/tensorflow that referenced this pull request Jan 26, 2026
…ntel GPU

Imported from GitHub PR openxla/xla#35304

This PR will enable sycl/oneapi plugin support for Intel GPU
Copybara import of the project:

--
cca65a21e9e59c38e961ccd68b036a2c52403728 by Ashiq Imran <ashiq.imran@intel.com>:

Adding sycl/oneapi plugin support

--
a0a9db1abcd5b84fc503fdc6a2502adeebf813d4 by Ashiq Imran <ashiq.imran@intel.com>:

addressing review comments

--
8d2f1dd19d4e6ba9d521003b4f047e687feffd3d by ag.ramesh <ag.ramesh@intel.com>:

Added if_cuda_is_configured check that was removed and reverted incorrect change.

Merging this change closes #35304

FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#35304 from Intel-tensorflow:aimran/oneapi_plugin 8d2f1dd19d4e6ba9d521003b4f047e687feffd3d
PiperOrigin-RevId: 853719463
@ashiqimranintel
Copy link
Contributor Author

@penpornk , @beckerhe , I resolved merge conflict, can you check it again?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

kokoro:force-run Forces CI to rerun

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants