[ROCm] Enable expandable segments by pragupta · Pull Request #173330 · pytorch/pytorch

pragupta · 2026-01-25T23:01:27Z

cc @jeffdaily @sunway513 @jithunnair-amd @pruthvistony @ROCmSupport @jataylo @hongxiayang @naromero77amd @jerrymannil @xinyazhang @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @kadeng @chauhang @amjames @Lucaskabela

pytorch-bot · 2026-01-25T23:01:31Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/173330

📄 Preview Python docs built from this PR
📄 Preview C++ docs built from this PR
❓ Need help or want to give feedback on the CI? Visit the bot commands wiki

Note: Links to docs will display an error until the docs builds have been completed.

❌ 1 New Failure, 13 Unrelated Failures

As of commit 7a4a90a with merge base 8be2451 ():

NEW FAILURE - The following job has failed:

trunk / linux-jammy-cuda13.0-py3.10-gcc11 / test (default, 4, 5, lf.linux.g6.4xlarge.experimental.nvidia.gpu) (gh)
test/dynamo/test_aot_compile.py::TestAOTCompile::test_aot_compile_with_aoti

FLAKY - The following jobs failed but were likely due to flakiness present on trunk:

inductor / unit-test / inductor-test / test (inductor, 1, 2, linux.g5.4xlarge.nvidia.gpu) (gh) (similar failure)
test/inductor/test_torchinductor_opinfo.py::TestInductorOpInfoCUDA::test_comprehensive_pca_lowrank_cuda_float32
inductor / unit-test / inductor-test / test (inductor, 2, 2, linux.g5.4xlarge.nvidia.gpu) (gh) (disabled by #137684)
test/inductor/test_torchinductor_opinfo.py::TestInductorOpInfoCUDA::test_comprehensive_linalg_lu_cuda_float32
pull / linux-jammy-py3.10-clang15 / test (dynamo_wrapped, 1, 3, lf.linux.2xlarge) (gh) (similar failure)
test/test_overrides.py::TestTorchFunctionMode::test_disable_subclass_mode
pull / linux-jammy-py3.14-clang15 / test (dynamo_wrapped, 1, 3, lf.linux.2xlarge) (gh) (similar failure)
test/test_overrides.py::TestTorchFunctionMode::test_disable_subclass_mode
pull / linux-jammy-py3.14t-clang15 / test (dynamo_wrapped, 1, 3, lf.linux.2xlarge) (gh) (similar failure)
test/test_overrides.py::TestTorchFunctionMode::test_disable_subclass_mode
pull / linux-jammy-py3.14t-clang15 / test (dynamo_wrapped, 2, 3, lf.linux.2xlarge) (gh) (similar failure)
Process completed with exit code 137.
rocm-mi300 / linux-noble-rocm-py3.12-mi300 / test (default, 3, 6, linux.rocm.gpu.gfx942.1) (gh) (detected as infra flaky with no log or failing log classifier)

BROKEN TRUNK - The following jobs failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

pull / linux-jammy-py3.10-clang15 / test (crossref, 2, 2, lf.linux.2xlarge) (gh) (trunk failure)
test/test_dynamic_shapes.py::TestUbackedOps::test_shape_id_runtime_assertion_on_mismatch
pull / linux-jammy-py3.14-clang15 / test (crossref, 2, 2, lf.linux.2xlarge) (gh) (trunk failure)
test/test_dynamic_shapes.py::TestUbackedOps::test_shape_id_runtime_assertion_on_mismatch
pull / linux-jammy-py3.14t-clang15 / test (crossref, 2, 2, lf.linux.2xlarge) (gh) (trunk failure)
test/test_dynamic_shapes.py::TestUbackedOps::test_shape_id_runtime_assertion_on_mismatch
torchtitan-test / torchtitan-x-pytorch-test / test (torchtitan_features_integration, 1, 1, linux.g5.48xlarge.nvidia.gpu) (gh) (trunk failure)
RuntimeError: 1 test steps failed: ['scripts/ci/pytorch_ci_test_runner.sh feature_tests']

UNSTABLE - The following jobs are marked as unstable, possibly due to flakiness on trunk:

inductor / inductor-cpu-test / test (cpu_inductor_torchbench, 1, 2, linux.2xlarge.amx, unstable) (gh) (#174929)
detectron2_maskrcnn_r_50_fpn
trunk / linux-jammy-rocm-py3.10 / test (distributed, 1, 3, linux.rocm.gpu.gfx950.4, unstable) (gh) (#177301)
'Test'

This comment was automatically generated by Dr. CI and updates every 15 minutes.

github-actions · 2026-01-25T23:03:10Z

This PR needs a `release notes:` label

If your changes are user facing and intended to be a part of release notes, please use a label starting with release notes:.

If not, please add the topic: not user facing label.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "topic: not user facing"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

linux-foundation-easycla · 2026-01-26T12:09:26Z

The committers listed above are authorized under a signed CLA.

✅ login: jeffdaily / name: Jeff Daily (10f8f71, 39ac3a0, 3cb4d19, 48c1178, 6b4dfb5, 9fe2ff3, ab6f10e, c5d30d6, f7b2ae9)
✅ login: moonshadow-25 / name: MoonShadow (0f7da95)
✅ login: pragupta / name: Prachi Gupta (ab0298a)
✅ login: Vighaneshs / name: Vighanesh Sharma (7a4a90a)

jeffdaily · 2026-02-03T18:24:14Z

We have found that for unit tests to fully pass we need this HIP patch ROCm/rocm-systems#3023.

jeffdaily · 2026-02-03T22:25:01Z

@pytorchbot rebase

pytorchmergebot · 2026-02-03T22:26:40Z

@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here

pytorchmergebot · 2026-02-03T22:26:45Z

Successfully rebased rocm_expandable_segments onto refs/remotes/origin/viable/strict, please pull locally before adding more changes (for example, via git checkout rocm_expandable_segments && git pull --rebase)

jeffdaily · 2026-02-20T02:57:35Z

@pytorchbot rebase

pytorchmergebot · 2026-02-20T02:59:16Z

@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here

pytorchmergebot · 2026-02-20T02:59:20Z

Successfully rebased rocm_expandable_segments onto refs/remotes/origin/viable/strict, please pull locally before adding more changes (for example, via git checkout rocm_expandable_segments && git pull --rebase)

jeffdaily · 2026-02-25T19:49:24Z

@pytorchbot merge

pytorch-bot · 2026-02-25T19:49:28Z

This PR needs to be approved by an authorized maintainer before merge.

jeffdaily · 2026-03-12T17:42:53Z

Noting that the 1 current failure seen is also seen on other PRs so it's not related.

https://hud.pytorch.org/failure?name=trunk%20%2F%20linux-jammy-cuda13.0-py3.10-gcc11%20%2F%20test%20(default%2C%204%2C%205%2C%20lf.linux.g6.4xlarge.experimental.nvidia.gpu)&jobName=linux-jammy-cuda13.0-py3.10-gcc11%20%2F%20test%20(default%2C%204%2C%205%2C%20lf.linux.g6.4xlarge.experimental.nvidia.gpu)&failureCaptures=test%2Fdynamo%2Ftest_aot_compile.py%3A%3ATestAOTCompile%3A%3Atest_aot_compile_with_aoti

jeffdaily · 2026-03-13T23:26:46Z

@pytorchbot merge -f "need to use force merge due to unrelated blocking failure, all other flaky CI is known; reason for revert has been addressed"

pytorchmergebot · 2026-03-13T23:28:37Z

Merge started

Your change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use -f as last resort and instead consider -i/--ignore-current to continue the merge ignoring current failures. This will allow currently pending tests to finish and report signal before the merge.

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here

yangw-dev · 2026-03-19T18:01:50Z

@pytorchbot revert -m "reverted internally, original:D96556656, revert diff: D96725665" -c ghfirst

pytorchmergebot · 2026-03-19T18:03:51Z

@pytorchbot successfully started a revert job. Check the current status here.
Questions? Feedback? Please reach out to the PyTorch DevX Team

This reverts commit 088c5a7. Reverted #173330 on behalf of https://github.com/yangw-dev due to reverted internally, original:D96556656, revert diff: D96725665 ([comment](#173330 (comment)))

pytorchmergebot · 2026-03-19T18:04:20Z

@pragupta your PR has been successfully reverted.

huydhn · 2026-03-19T18:36:42Z

Let me import to reland this

meta-codesync · 2026-03-19T18:40:38Z

@huydhn has imported this pull request. If you are a Meta employee, you can view this in D97339294.

## Motivation Fixes #3962 - The `rocprofiler-sdk` shared library is not being preloaded, causing `librocprofiler-sdk.so.1` to be missing at runtime. This is because the PyTorch `kineto` submodule was bumped which switched from `roctracer` to `rocprofiler-sdk`: pytorch/pytorch#177101 - `test_mempool_expandable` was enabled on ROCm by pytorch/pytorch#173330. This test was failing as it requires the rocm[devel] packages but was causing a crash: https://github.com/ROCm/TheRock/actions/runs/23164829934/job/67321547840. This test is currently already skipped for other torch versions. - Also skip `test_mempool_empty_cache_inactive`, `test_mempool_limited_memory_with_allocator`, `test_deleted_mempool_not_used_on_oom`, and `test_mempool_ctx_multithread` as these also require building `dummy_allocator` and are skipped in other torch versions. ## Technical Details - Adds `rocprofiler-sdk` to `LINUX_LIBRARY_PRELOADS` in `build_prod_wheels.py` so that `librocprofiler-sdk.so` is loaded - Registers `rocprofiler-sdk` as a `LibraryEntry` in `_dist_info.py` so the `rocm_sdk` package can resolve the name to the actual `.so` file. ## Test Plan - Verify that ROCm builds, the nightly smoke tests pass and that running the torch tests do not crash ## Test Result - ROCm builds successfully: https://github.com/ROCm/TheRock/actions/runs/23152017500 - Smoke tests pass for torch nightly and the runner is not crashing: https://github.com/ROCm/TheRock/actions/runs/23253453219 ## Submission Checklist - [x] Look over the contributing guidelines at https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.

Summary: Original pull request: #173330 Fixes #168737. Fixes #168736. The original diff enabled expandable segments for ROCm by adding `#ifdef USE_ROCM` guards throughout CUDACachingAllocator.cpp to use HIP APIs (hipMemAddressReserve, hipMemCreate, hipMemMap, etc.) instead of CUDA driver APIs when building for ROCm. Root cause: In HIP/ROCm 6.2.1, the field name for memory allocation properties is `requestedHandleType` (singular), not `requestedHandleTypes` (plural) as in CUDA. Additionally, `hipMemHandleTypeFabric` does not exist in HIP, so the `CU_MEM_HANDLE_TYPE_FABRIC` assignment must be skipped on ROCm. Fix applied on top of the original diff (from D96652342): - Use `prop.requestedHandleType = hipMemHandleTypePosixFileDescriptor` under `#ifdef USE_ROCM` (singular field name, HIP constant) - Use `prop.requestedHandleTypes = CU_MEM_HANDLE_TYPE_POSIX_FILE_DESCRIPTOR` for CUDA (plural field name, CUDA constant) - Skip the `CU_MEM_HANDLE_TYPE_FABRIC` assignment entirely on ROCm under `#ifndef USE_ROCM`, as `hipMemHandleTypeFabric` does not exist in HIP Co-authored-by: Prachi Gupta prachi.gupta@amd.com Co-authored-by: Jeff Daily jeff.daily@amd.com Co-authored-by: moonshadow-25 moonshadow-25@users.noreply.github.com Co-authored-by: Vighanesh Sharma vighaneshsharma@gmail.com Test Plan: ``` fbpkg build //aps_models/ads/ecosystem/eval/cogwheel_tests/amd:cogwheel_aps_ads_icvr_kd_eval_amd_test_harness --build-remote ``` https://www.internalfb.com/sandcastle/workflow/1049338713192153464 Differential Revision: D97211385

Summary: Original pull request: #173330 Fixes #168737. Fixes #168736. The original diff enabled expandable segments for ROCm by adding `#ifdef USE_ROCM` guards throughout CUDACachingAllocator.cpp to use HIP APIs (hipMemAddressReserve, hipMemCreate, hipMemMap, etc.) instead of CUDA driver APIs when building for ROCm. Root cause: In HIP/ROCm 6.2.1, the field name for memory allocation properties is `requestedHandleType` (singular), not `requestedHandleTypes` (plural) as in CUDA. Additionally, `hipMemHandleTypeFabric` does not exist in HIP, so the `CU_MEM_HANDLE_TYPE_FABRIC` assignment must be skipped on ROCm. Fix applied on top of the original diff (from D96652342): - Use `prop.requestedHandleType = hipMemHandleTypePosixFileDescriptor` under `#ifdef USE_ROCM` (singular field name, HIP constant) - Use `prop.requestedHandleTypes = CU_MEM_HANDLE_TYPE_POSIX_FILE_DESCRIPTOR` for CUDA (plural field name, CUDA constant) - Skip the `CU_MEM_HANDLE_TYPE_FABRIC` assignment entirely on ROCm under `#ifndef USE_ROCM`, as `hipMemHandleTypeFabric` does not exist in HIP Co-authored-by: Prachi Gupta prachi.gupta@amd.com Co-authored-by: Jeff Daily jeff.daily@amd.com Co-authored-by: moonshadow-25 moonshadow-25@users.noreply.github.com Co-authored-by: Vighanesh Sharma vighaneshsharma@gmail.com Test Plan: ``` fbpkg build //aps_models/ads/ecosystem/eval/cogwheel_tests/amd:cogwheel_aps_ads_icvr_kd_eval_amd_test_harness --build-remote ``` https://www.internalfb.com/sandcastle/workflow/1049338713192153464 Differential Revision: D97211385 Pull Request resolved: #177974 Approved by: https://github.com/jeffdaily, https://github.com/echen4096

## Motivation Fixes #3962 - The `rocprofiler-sdk` shared library is not being preloaded, causing `librocprofiler-sdk.so.1` to be missing at runtime. This is because the PyTorch `kineto` submodule was bumped which switched from `roctracer` to `rocprofiler-sdk`: pytorch/pytorch#177101 - `test_mempool_expandable` was enabled on ROCm by pytorch/pytorch#173330. This test was failing as it requires the rocm[devel] packages but was causing a crash: https://github.com/ROCm/TheRock/actions/runs/23164829934/job/67321547840. This test is currently already skipped for other torch versions. - Also skip `test_mempool_empty_cache_inactive`, `test_mempool_limited_memory_with_allocator`, `test_deleted_mempool_not_used_on_oom`, and `test_mempool_ctx_multithread` as these also require building `dummy_allocator` and are skipped in other torch versions. ## Technical Details - Adds `rocprofiler-sdk` to `LINUX_LIBRARY_PRELOADS` in `build_prod_wheels.py` so that `librocprofiler-sdk.so` is loaded - Registers `rocprofiler-sdk` as a `LibraryEntry` in `_dist_info.py` so the `rocm_sdk` package can resolve the name to the actual `.so` file. ## Test Plan - Verify that ROCm builds, the nightly smoke tests pass and that running the torch tests do not crash ## Test Result - ROCm builds successfully: https://github.com/ROCm/TheRock/actions/runs/23152017500 - Smoke tests pass for torch nightly and the runner is not crashing: https://github.com/ROCm/TheRock/actions/runs/23253453219 ## Submission Checklist - [x] Look over the contributing guidelines at https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.

pragupta · 2026-03-24T15:41:40Z

closing this one as relanded here: #177974

…77974) Summary: Original pull request: pytorch#173330 Fixes pytorch#168737. Fixes pytorch#168736. The original diff enabled expandable segments for ROCm by adding `#ifdef USE_ROCM` guards throughout CUDACachingAllocator.cpp to use HIP APIs (hipMemAddressReserve, hipMemCreate, hipMemMap, etc.) instead of CUDA driver APIs when building for ROCm. Root cause: In HIP/ROCm 6.2.1, the field name for memory allocation properties is `requestedHandleType` (singular), not `requestedHandleTypes` (plural) as in CUDA. Additionally, `hipMemHandleTypeFabric` does not exist in HIP, so the `CU_MEM_HANDLE_TYPE_FABRIC` assignment must be skipped on ROCm. Fix applied on top of the original diff (from D96652342): - Use `prop.requestedHandleType = hipMemHandleTypePosixFileDescriptor` under `#ifdef USE_ROCM` (singular field name, HIP constant) - Use `prop.requestedHandleTypes = CU_MEM_HANDLE_TYPE_POSIX_FILE_DESCRIPTOR` for CUDA (plural field name, CUDA constant) - Skip the `CU_MEM_HANDLE_TYPE_FABRIC` assignment entirely on ROCm under `#ifndef USE_ROCM`, as `hipMemHandleTypeFabric` does not exist in HIP Co-authored-by: Prachi Gupta prachi.gupta@amd.com Co-authored-by: Jeff Daily jeff.daily@amd.com Co-authored-by: moonshadow-25 moonshadow-25@users.noreply.github.com Co-authored-by: Vighanesh Sharma vighaneshsharma@gmail.com Test Plan: ``` fbpkg build //aps_models/ads/ecosystem/eval/cogwheel_tests/amd:cogwheel_aps_ads_icvr_kd_eval_amd_test_harness --build-remote ``` https://www.internalfb.com/sandcastle/workflow/1049338713192153464 Differential Revision: D97211385 Pull Request resolved: pytorch#177974 Approved by: https://github.com/jeffdaily, https://github.com/echen4096 (cherry picked from commit 5792701)

…77974) (#3106) Summary: Original pull request: pytorch#173330 Fixes pytorch#168737. Fixes pytorch#168736. The original diff enabled expandable segments for ROCm by adding `#ifdef USE_ROCM` guards throughout CUDACachingAllocator.cpp to use HIP APIs (hipMemAddressReserve, hipMemCreate, hipMemMap, etc.) instead of CUDA driver APIs when building for ROCm. Root cause: In HIP/ROCm 6.2.1, the field name for memory allocation properties is `requestedHandleType` (singular), not `requestedHandleTypes` (plural) as in CUDA. Additionally, `hipMemHandleTypeFabric` does not exist in HIP, so the `CU_MEM_HANDLE_TYPE_FABRIC` assignment must be skipped on ROCm. Fix applied on top of the original diff (from D96652342): - Use `prop.requestedHandleType = hipMemHandleTypePosixFileDescriptor` under `#ifdef USE_ROCM` (singular field name, HIP constant) - Use `prop.requestedHandleTypes = CU_MEM_HANDLE_TYPE_POSIX_FILE_DESCRIPTOR` for CUDA (plural field name, CUDA constant) - Skip the `CU_MEM_HANDLE_TYPE_FABRIC` assignment entirely on ROCm under `#ifndef USE_ROCM`, as `hipMemHandleTypeFabric` does not exist in HIP Co-authored-by: Prachi Gupta prachi.gupta@amd.com Co-authored-by: Jeff Daily jeff.daily@amd.com Co-authored-by: moonshadow-25 moonshadow-25@users.noreply.github.com Co-authored-by: Vighanesh Sharma vighaneshsharma@gmail.com Test Plan: ``` fbpkg build //aps_models/ads/ecosystem/eval/cogwheel_tests/amd:cogwheel_aps_ads_icvr_kd_eval_amd_test_harness --build-remote ``` https://www.internalfb.com/sandcastle/workflow/1049338713192153464 Differential Revision: D97211385 Pull Request resolved: pytorch#177974 Approved by: https://github.com/jeffdaily, https://github.com/echen4096 (cherry picked from commit 5792701) ## Motivation  ## Technical Details  ## Test Plan  ## Test Result  ## Submission Checklist - [ ] Look over the contributing guidelines at https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests. Co-authored-by: Haoyu Zhang <haoyuz@meta.com>

…77974) Summary: Original pull request: pytorch#173330 Fixes pytorch#168737. Fixes pytorch#168736. The original diff enabled expandable segments for ROCm by adding `#ifdef USE_ROCM` guards throughout CUDACachingAllocator.cpp to use HIP APIs (hipMemAddressReserve, hipMemCreate, hipMemMap, etc.) instead of CUDA driver APIs when building for ROCm. Root cause: In HIP/ROCm 6.2.1, the field name for memory allocation properties is `requestedHandleType` (singular), not `requestedHandleTypes` (plural) as in CUDA. Additionally, `hipMemHandleTypeFabric` does not exist in HIP, so the `CU_MEM_HANDLE_TYPE_FABRIC` assignment must be skipped on ROCm. Fix applied on top of the original diff (from D96652342): - Use `prop.requestedHandleType = hipMemHandleTypePosixFileDescriptor` under `#ifdef USE_ROCM` (singular field name, HIP constant) - Use `prop.requestedHandleTypes = CU_MEM_HANDLE_TYPE_POSIX_FILE_DESCRIPTOR` for CUDA (plural field name, CUDA constant) - Skip the `CU_MEM_HANDLE_TYPE_FABRIC` assignment entirely on ROCm under `#ifndef USE_ROCM`, as `hipMemHandleTypeFabric` does not exist in HIP Co-authored-by: Prachi Gupta prachi.gupta@amd.com Co-authored-by: Jeff Daily jeff.daily@amd.com Co-authored-by: moonshadow-25 moonshadow-25@users.noreply.github.com Co-authored-by: Vighanesh Sharma vighaneshsharma@gmail.com Test Plan: ``` fbpkg build //aps_models/ads/ecosystem/eval/cogwheel_tests/amd:cogwheel_aps_ads_icvr_kd_eval_amd_test_harness --build-remote ``` https://www.internalfb.com/sandcastle/workflow/1049338713192153464 Differential Revision: D97211385 Pull Request resolved: pytorch#177974 Approved by: https://github.com/jeffdaily, https://github.com/echen4096

Pull Request resolved: pytorch#173330 Approved by: https://github.com/jeffdaily Co-authored-by: Jeff Daily <jeff.daily@amd.com>

pragupta requested review from Aidyn-A, eqy and syed-ahmed as code owners January 25, 2026 23:01

pytorch-bot bot added ciflow/rocm-mi300 Trigger "default" config CI on ROCm MI300 module: rocm AMD GPU support for Pytorch labels Jan 25, 2026

pragupta added the ciflow/trunk Trigger trunk jobs on your pull request label Jan 25, 2026

pragupta mentioned this pull request Jan 25, 2026

[ROCm] Enable expandable segments #169068

Closed

pytorchbot added the open source label Jan 25, 2026

pragupta force-pushed the rocm_expandable_segments branch from b19983c to 3b3ffe3 Compare January 26, 2026 12:09

pragupta force-pushed the rocm_expandable_segments branch from 3b3ffe3 to e4e0d36 Compare January 26, 2026 12:09

bdhirsh added the triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module label Jan 26, 2026

pragupta marked this pull request as draft January 27, 2026 15:20

jeffdaily added the release notes: rocm mandatorylabel label Feb 3, 2026

pytorchmergebot force-pushed the rocm_expandable_segments branch from e4e0d36 to d6fed57 Compare February 3, 2026 22:26

jeffdaily marked this pull request as ready for review February 11, 2026 16:38

pytorch-bot bot added ciflow/inductor module: dynamo labels Feb 12, 2026

pytorchmergebot force-pushed the rocm_expandable_segments branch from 78869d4 to 922a0d4 Compare February 20, 2026 02:59

pytorchmergebot added the merging label Mar 13, 2026

pytorchmergebot closed this in 088c5a7 Mar 13, 2026

pytorchmergebot removed the merging label Mar 13, 2026

huydhn mentioned this pull request Mar 16, 2026

[On-call] Fix MI300X conveyor build failure (#177503) #177503

Open

darren-amd mentioned this pull request Mar 19, 2026

[torch] Preload rocprofiler-sdk to fix nightly smoketests ROCm/TheRock#4065

Merged

1 task

pytorchmergebot reopened this Mar 19, 2026

haoyuz mentioned this pull request Mar 20, 2026

[ROCm] Reland: Enable expandable segments (#173330) #177974

Closed

pragupta closed this Mar 24, 2026

pragupta mentioned this pull request Mar 26, 2026

[release/2.11][ROCm] Reland: Enable expandable segments (#173330) (#177974) ROCm/pytorch#3106

Merged

1 task

EmanueleCoradin pushed a commit to EmanueleCoradin/pytorch that referenced this pull request Mar 30, 2026

[ROCm] Enable expandable segments (pytorch#173330)

010dab6

Pull Request resolved: pytorch#173330 Approved by: https://github.com/jeffdaily Co-authored-by: Jeff Daily <jeff.daily@amd.com>

Conversation

pragupta commented Jan 25, 2026 • edited by jeffdaily Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-bot bot commented Jan 25, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/173330

❌ 1 New Failure, 13 Unrelated Failures

Uh oh!

github-actions bot commented Jan 25, 2026

This PR needs a release notes: label

Uh oh!

linux-foundation-easycla bot commented Jan 26, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

jeffdaily commented Feb 3, 2026

Uh oh!

jeffdaily commented Feb 3, 2026

Uh oh!

pytorchmergebot commented Feb 3, 2026

Uh oh!

pytorchmergebot commented Feb 3, 2026

Uh oh!

jeffdaily commented Feb 20, 2026

Uh oh!

pytorchmergebot commented Feb 20, 2026

Uh oh!

pytorchmergebot commented Feb 20, 2026

Uh oh!

jeffdaily commented Feb 25, 2026

Uh oh!

pytorch-bot bot commented Feb 25, 2026

Uh oh!

jeffdaily commented Mar 12, 2026

Uh oh!

jeffdaily commented Mar 13, 2026

Uh oh!

pytorchmergebot commented Mar 13, 2026

Merge started

Uh oh!

yangw-dev commented Mar 19, 2026

Uh oh!

pytorchmergebot commented Mar 19, 2026

Uh oh!

pytorchmergebot commented Mar 19, 2026

Uh oh!

huydhn commented Mar 19, 2026

Uh oh!

meta-codesync bot commented Mar 19, 2026

Uh oh!

pragupta commented Mar 24, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

12 participants

pragupta commented Jan 25, 2026 •

edited by jeffdaily

Loading

pytorch-bot bot commented Jan 25, 2026 •

edited

Loading

This PR needs a `release notes:` label

linux-foundation-easycla bot commented Jan 26, 2026 •

edited

Loading