Maintainer Moves

The row is the compact decision object: move, why, next action, affected scope, confidence, and compact evidence. Expand a row to inspect its bundles; review threads and raw public evidence rows stay underneath for audit.

workflow compression
9
Maintainer moves
same next action grouped
15437
Issues covered
open issues
2770
PRs covered
open PRs
2704
Review groups
expandable underneath
99% fewer than review groups
Compression
workflow groups
0
GitHub writes
local-only; no GitHub writes

Recommended Work Plan

Ordered by maintainer workflow lane, then by attention score. Each row expands from action summary to bundle evidence without exposing raw API payloads.

make final PR decision

Maintainer-owned unblock queue
Why

The local packet found no failed-check blocker; the next step is maintainer review, merge, close, rebase request, or takeover.

Next action

Review changed files, linked issues, and latest comments, then choose merge, close, rebase request, or maintainer takeover.

Impact

329 bundles | 259 issues | 988 PRs
Confidence: high

Evidence

failed CI checksno failed-check blockerreviewDecision=APPROVEDmergeStateStatus=UNKNOWN329 package receipts

View bundles

Bundle BUNDLE-002: PR #189122

Existing PR decisionwaiting on maintainer review2 issues1 PRmedium
Issues #158861#188900
PRs PR #189122
  • Diagnosis: PR #189122 has no local failed-check diagnosis; it needs a maintainer merge/close/rebase decision.
  • Next action: Review changed files, linked issues, and latest maintainer comments, then choose merge, close, rebase request, or maintainer takeover.
  • Evidence: reviewDecision=APPROVED; mergeStateStatus=UNKNOWN; pending=; stale_days=1
  • Short excerpt: pr #189122 declares a closing reference to issue #188900. | Fixes #188900 Summary Multiplying a dense tensor by a sparse COO tensor whose size-1 sparse dimension must broadcast up to a larger size silently dropped | Looks correct | s-is, so only the slice at index 0 along the...
  • Review threads: thread:issue:158861:issue:188900:pr:189122
  • Package receipt: 7c283f0cf9f9

Bundle BUNDLE-005: PR #181720 / PR #189256

Existing PR decisionwaiting on maintainer review3 issues2 PRsmedium
Issues #180397#181374#188970
PRs PR #181720PR #189256
  • Diagnosis: PR #189256 has no local failed-check diagnosis; it needs a maintainer merge/close/rebase decision.
  • Next action: Review changed files, linked issues, and latest maintainer comments, then choose merge, close, rebase request, or maintainer takeover.
  • Evidence: reviewDecision=APPROVED; mergeStateStatus=CLEAN; pending=; stale_days=0
  • Short excerpt: at in the current form it gives no perf benefit over regular CPU memory, it this is a foundation for copy-on-write work Fixes #181374 Fixes #188970 Authored with assistance from Claude Code. | Please note that in the current form it gives no perf benefit over regular CPU memor...
  • Review threads: thread:issue:180397:issue:181374:issue:188970:pr:181720
  • Package receipt: b3f507176f68

Bundle BUNDLE-011: PR #189045

Existing PR decisionwaiting on maintainer review3 issues1 PRmedium
Issues #40932#41508#55056
PRs PR #189045
  • Diagnosis: PR #189045 has no local failed-check diagnosis; it needs a maintainer merge/close/rebase decision.
  • Next action: Review changed files, linked issues, and latest maintainer comments, then choose merge, close, rebase request, or maintainer takeover.
  • Evidence: reviewDecision=; mergeStateStatus=UNKNOWN; pending=; stale_days=0
  • Short excerpt: pr #189045 declares a closing reference to issue #41508. | ---Fixes #41508 Root cause nn.MultiheadAttention produces NaN gradients when a key_padding_mask (or attn_mask) fully masks out every key for a given query | Hi @HelenaHlz, as directed by @jbschlosser please refer to th...
  • Review threads: thread:issue:40932:issue:41508:issue:55056:pr:189045
  • Package receipt: e38e5b031d03

Bundle BUNDLE-015: PR #187354 / PR #188556

Existing PR decisionwaiting on maintainer review2 issues2 PRsmedium
Issues #187332#188545
PRs PR #187354PR #188556
  • Diagnosis: PR #187354 has no local failed-check diagnosis; it needs a maintainer merge/close/rebase decision.
  • Next action: Review changed files, linked issues, and latest maintainer comments, then choose merge, close, rebase request, or maintainer takeover.
  • Evidence: reviewDecision=APPROVED; mergeStateStatus=CLEAN; pending=; stale_days=0
  • Short excerpt: pr #188556 declares a closing reference to issue #188545. | Fixes #188545 Summary Preserve eager-mode semantics for infinite inputs in the Triton lowering of torch.special.bessel_j0 torch.special.bessel_j1 torch.s | ile continuing to use libdevice for finite inputs. This prese...
  • Review threads: thread:issue:187332:issue:188545:pr:187354:pr:188556
  • Package receipt: 0c1a21954532

Bundle BUNDLE-022: PR #189273

Existing PR decisionwaiting on CI/check fix1 issue1 PRhigh
Issues #189267
PRs PR #189273
  • Diagnosis: PR #189273 is blocked by failing CI checks: statusCheckRollup.
  • Next action: Fix or ask the contributor to fix statusCheckRollup, rerun checks, then revisit review/owner-gate approval once the branch is green.
  • Evidence: failed checks: statusCheckRollup.
  • Short excerpt: See #189267 (comment) | pr #189273 declares a closing reference to issue #189267. | Fixes #189267 | PR #189273: draft/noise; checks: statusCheckRollup, statusCheckRollup; review: APPROVED
  • Review threads: thread:issue:189267:pr:189273
  • Package receipt: 4b094aa505ba

Bundle BUNDLE-025: PR #188791

Existing PR decisionwaiting on maintainer review1 issue1 PRmedium
Issues #188792
PRs PR #188791
  • Diagnosis: PR #188791 has no local failed-check diagnosis; it needs a maintainer merge/close/rebase decision.
  • Next action: Review changed files, linked issues, and latest maintainer comments, then choose merge, close, rebase request, or maintainer takeover.
  • Evidence: reviewDecision=; mergeStateStatus=UNKNOWN; pending=; stale_days=4
  • Short excerpt: each worker do a serial 1-element torch.sqrt first; this drops the rate to 0 (5,200 first-calls), which is the basis of the fix proposed in #188791. mkl_vml_race_repro.py """Minimal repro: MKL VML first-call dispatch race in torch.sqrt. Spawns waves of worker processes on an o...
  • Review threads: thread:issue:188792:pr:188791
  • Package receipt: 80accb61c397

Bundle BUNDLE-028: PR #188114

Existing PR decisionwaiting on maintainer review1 issue1 PRmedium
Issues #188113
PRs PR #188114
  • Diagnosis: PR #188114 has no local failed-check diagnosis; it needs a maintainer merge/close/rebase decision.
  • Next action: Review changed files, linked issues, and latest maintainer comments, then choose merge, close, rebase request, or maintainer takeover.
  • Evidence: reviewDecision=APPROVED; mergeStateStatus=UNKNOWN; pending=; stale_days=1
  • Short excerpt: pr #188114 declares a closing reference to issue #188113. | x950. Added gfx1200/gfx1201. Tested on gfx1201 (RX 9070): WITHOUT patch the kernel ASSERTs, WITH patch the kernel executes correctly. Fixes #188113 cc @jeffdaily @sunway513 @jithunnair-amd @pruthvistony @ROCmSupport @...
  • Review threads: thread:issue:188113:pr:188114
  • Package receipt: f4e4370f839a

Bundle BUNDLE-030: PR #188039

Existing PR decisionwaiting on maintainer review1 issue1 PRmedium
Issues #188034
PRs PR #188039
  • Diagnosis: PR #188039 has no local failed-check diagnosis; it needs a maintainer merge/close/rebase decision.
  • Next action: Review changed files, linked issues, and latest maintainer comments, then choose merge, close, rebase request, or maintainer takeover.
  • Evidence: reviewDecision=; mergeStateStatus=UNKNOWN; pending=; stale_days=1
  • Short excerpt: pr #188039 declares a closing reference to issue #188034. | st that returns a sentinel object from setup_context and asserts sys.getrefcount is unchanged after repeated forward+backward passes. Fixes #188034 Test plan python test/test_autograd.py TestAutograd.test_custom_funct...
  • Review threads: thread:issue:188034:pr:188039
  • Package receipt: 36000e9d2f4b

Bundle BUNDLE-035: PR #187104

Existing PR decisionwaiting on maintainer review1 issue1 PRmedium
Issues #175370
PRs PR #187104
  • Diagnosis: PR #187104 has no local failed-check diagnosis; it needs a maintainer merge/close/rebase decision.
  • Next action: Review changed files, linked issues, and latest maintainer comments, then choose merge, close, rebase request, or maintainer takeover.
  • Evidence: reviewDecision=; mergeStateStatus=UNKNOWN; pending=; stale_days=1
  • Short excerpt: pr #187104 declares a closing reference to issue #175370. | Fixes #175370 F.embedding_bag on CPU segfaults when offsets is empty and indices is not. Root cause With zero bags, make_offset2bag fills offset2bag with | PR #187104: has linked issue; checks:; review:
  • Review threads: thread:issue:175370:pr:187104
  • Package receipt: 9ffed6bf077a

Bundle BUNDLE-037: PR #186583

Existing PR decisionwaiting on maintainer review1 issue1 PRmedium
Issues #186537
PRs PR #186583
  • Diagnosis: PR #186583 has no local failed-check diagnosis; it needs a maintainer merge/close/rebase decision.
  • Next action: Review changed files, linked issues, and latest maintainer comments, then choose merge, close, rebase request, or maintainer takeover.
  • Evidence: reviewDecision=APPROVED; mergeStateStatus=CLEAN; pending=; stale_days=1
  • Short excerpt: shapes to exceed the default recompile limit, and asserts both the forward and recompute phases see the caller's raised Dynamo limit. Fixes #186537 Generated by my agent Benchmark Results: Post-fix Python autograd.Function backward microbenchmark, 20,000 backward calls per rep...
  • Review threads: thread:issue:186537:pr:186583
  • Package receipt: 19c96580e3b7

Bundle BUNDLE-038: PR #186454

Existing PR decisionwaiting on maintainer review1 issue1 PRmedium
Issues #124863
PRs PR #186454
  • Diagnosis: PR #186454 has no local failed-check diagnosis; it needs a maintainer merge/close/rebase decision.
  • Next action: Review changed files, linked issues, and latest maintainer comments, then choose merge, close, rebase request, or maintainer takeover.
  • Evidence: reviewDecision=; mergeStateStatus=CLEAN; pending=; stale_days=0
  • Short excerpt: ore removal, and warning during registration gives users a direct migration signal without changing dispatcher semantics immediately. Fixes #124863 Generated by my agent Test Plan: python test/test_custom_ops.py TestCustomOp.test_incorrect_schema_types TestCustomOp.test_is_ten...
  • Review threads: thread:issue:124863:pr:186454
  • Package receipt: 65bb42ece760

Bundle BUNDLE-046: PR #183701

Existing PR decisionwaiting on maintainer review1 issue1 PRmedium
Issues #181690
PRs PR #183701
  • Diagnosis: PR #183701 has no local failed-check diagnosis; it needs a maintainer merge/close/rebase decision.
  • Next action: Review changed files, linked issues, and latest maintainer comments, then choose merge, close, rebase request, or maintainer takeover.
  • Evidence: reviewDecision=; mergeStateStatus=CLEAN; pending=; stale_days=0
  • Short excerpt: duction suffix and emitted full-width non-contiguous stores. Reuse the tail vector kernel suffix so tail_size bounds the final store. Fixes #181690 Generated by my agent cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv...
  • Review threads: thread:issue:181690:pr:183701
  • Package receipt: d6aeb645e15d
317 more bundle receipt(s)

Remaining bundles stay in the downloadable public-safe CSV artifacts to keep this action tab fast to scan.

fix failing CI checks

Repair-ready PRs
Why

The PR is not ready for maintainer decision until the named CI failure is repaired.

Next action

Use the failed check and changed-file evidence to pick the smallest repair path, then rerun the failing job.

Impact

484 bundles | 711 issues | 1718 PRs
Confidence: high

Evidence

failed CI checksfailed checks: statusCheckRollup.failed checks: statusCheckRollup, unit-test / inductor-...failed checks: statusCheckRollup, inductor-test / test-...484 package receipts

View bundles

Bundle BUNDLE-003: PR #185730

Existing PR blocked reviewwaiting on CI/check fix2 issues1 PRhigh
Issues #150022#150371
PRs PR #185730
  • Diagnosis: PR #185730 is blocked by failing CI checks: statusCheckRollup.
  • Next action: Fix or ask the contributor to fix statusCheckRollup, rerun checks, then revisit review/owner-gate approval once the branch is green.
  • Evidence: failed checks: statusCheckRollup.
  • Short excerpt: the dynamic_shapes tree and the traced input tree had different structures; the trace-order remap fixes that path too. Fixes #150371 Fixes #150022 Generated by my agent Test Plan: python - <<'PY' ... PY issue repro for strict=False and strict=True: both exported successfully w...
  • Review threads: thread:issue:150022:issue:150371:pr:185730
  • Package receipt: a6fa3ccdc006

Bundle BUNDLE-004: PR #188948

Existing PR contributor waitwaiting on CI/check fix1 issue1 PRhigh
Issues #153410
PRs PR #188948
  • Diagnosis: PR #188948 is blocked by failing CI checks: statusCheckRollup.
  • Next action: Fix or ask the contributor to fix statusCheckRollup, rerun checks, then revisit review/owner-gate approval once the branch is green.
  • Evidence: failed checks: statusCheckRollup.
  • Short excerpt: pr #188948 declares a closing reference to issue #153410. | Fixes #153410 This change provides a reference implementation for triangular_solve on sparse CPU tensors when MKL is not available, by materializing diag | PR #188948: waiting on contributor; checks: statusCheckRollup...
  • Review threads: thread:issue:153410:pr:188948
  • Package receipt: 54ffdc39a6bd

Bundle BUNDLE-006: PR #189043

Existing PR contributor waitwaiting on CI/check fix1 issue1 PRhigh
Issues #188892
PRs PR #189043
  • Diagnosis: PR #189043 is blocked by failing CI checks: statusCheckRollup.
  • Next action: Fix or ask the contributor to fix statusCheckRollup, rerun checks, then revisit review/owner-gate approval once the branch is green.
  • Evidence: failed checks: statusCheckRollup.
  • Short excerpt: pr #189043 declares a closing reference to issue #188892. | Fixes #188892 Problem cuDNN 9 is split into a dispatcher (libcudnn.so) plus a set of engine sub-libraries (libcudnn_graph.so, libcudnn_engines_*.so, libc | PR #189043: waiting on contributor; checks: statusCheckRollup...
  • Review threads: thread:issue:188892:pr:189043
  • Package receipt: 808cc9baeb6f

Bundle BUNDLE-007: PR #181726 / PR #181727 / PR #181728 / PR #183511 / +10

Existing PR blocked reviewwaiting on CI/check fix21 issues14 PRshigh
Issues #178040#183988#186348#187988#188477#188675#188704#188706#188707#188708#188709#188710+9
PRs PR #181726PR #181727PR #181728PR #183511PR #184276PR #187315PR #187318PR #187989PR #188268PR #188931+4
  • Diagnosis: PR #189149 is blocked by failing CI checks: statusCheckRollup.
  • Next action: Fix or ask the contributor to fix statusCheckRollup, rerun checks, then revisit review/owner-gate approval once the branch is green.
  • Evidence: failed checks: statusCheckRollup.
  • Short excerpt: pr #189149 declares a closing reference to issue #186348. | pr #187989 declares a closing reference to issue #187988. | pr #188931 declares a closing reference to issue #188891. | pr #184276 declares a closing reference to issue #183988. | pr #188996 declares a closing referen...
  • Review threads: thread:issue:178040:issue:183988:issue:186348:issue:187988
  • Package receipt: 9a19487204ee

Bundle BUNDLE-008: PR #184481 / PR #184694 / PR #185891

Existing PR contributor waitwaiting on CI/check fix3 issues3 PRshigh
Issues #93617#178680#188133
PRs PR #184481PR #184694PR #185891
  • Diagnosis: PR #185891 is blocked by failing CI checks: statusCheckRollup.
  • Next action: Fix or ask the contributor to fix statusCheckRollup, rerun checks, then revisit review/owner-gate approval once the branch is green.
  • Evidence: failed checks: statusCheckRollup.
  • Short excerpt: including static traces and AOTAutograd cache hits, so alias-sensitive synthetic-base graphs are not reused for incompatible inputs. Fixes #178680 Generated by my agent cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv....
  • Review threads: thread:issue:178680:issue:188133:issue:93617:pr:184481
  • Package receipt: 3d3d3ef1fb3b

Bundle BUNDLE-009: PR #185946 / PR #186375

Existing PR blocked reviewwaiting on CI/check fix3 issues2 PRshigh
Issues #16706#185240#186374
PRs PR #185946PR #186375
  • Diagnosis: PR #185946 is blocked by failing CI checks: statusCheckRollup.
  • Next action: Fix or ask the contributor to fix statusCheckRollup, rerun checks, then revisit review/owner-gate approval once the branch is green.
  • Evidence: failed checks: statusCheckRollup.
  • Short excerpt: pr #186375 declares a closing reference to issue #186374. | pr #185946 declares a closing reference to issue #185240. | ad of falling back gracefully. Similar to #185946, this PR replaces the assertion with a warning and allows control flow to continue. Fixes #186374 cc @eqy |...
  • Review threads: thread:issue:16706:issue:185240:issue:186374:pr:185946
  • Package receipt: ba5b87b237c1

Bundle BUNDLE-010: PR #185846 / PR #186358 / PR #187609

Existing PR blocked reviewwaiting on CI/check fix2 issues3 PRshigh
Issues #185470#185484
PRs PR #185846PR #186358PR #187609
  • Diagnosis: PR #186358 is blocked by failing CI checks: statusCheckRollup.
  • Next action: Fix or ask the contributor to fix statusCheckRollup, rerun checks, then revisit review/owner-gate approval once the branch is green.
  • Evidence: failed checks: statusCheckRollup.
  • Short excerpt: pr #186358 declares a closing reference to issue #185484. | so output dtype semantics are unchanged. A prior stale PR, #185846, made the same root-cause fix but has not landed on current main. Fixes #185470 Generated by my agent Test Plan: python test/inductor/test_torchinduct...
  • Review threads: thread:issue:185470:issue:185484:pr:185846:pr:186358
  • Package receipt: 221d1417621c

Bundle BUNDLE-012: PR #188998

Existing PR blocked reviewwaiting on CI/check fix3 issues1 PRhigh
Issues #188492#188841#189287
PRs PR #188998
  • Diagnosis: PR #188998 is blocked by failing CI checks: statusCheckRollup.
  • Next action: Fix or ask the contributor to fix statusCheckRollup, rerun checks, then revisit review/owner-gate approval once the branch is green.
  • Evidence: failed checks: statusCheckRollup.
  • Short excerpt: pr #188998 declares a closing reference to issue #188492. | Fixes #188492 Under Triton 3.8 on Hopper (SM90), bf16 Swin-Transformer and similar bf16 workloads compiled with torch.compile(mode="max-autotune") showed | ibution needs confirmation (it is by failing-since alignment,...
  • Review threads: thread:issue:188492:issue:188841:issue:189287:pr:188998
  • Package receipt: e6447d3f086b

Bundle BUNDLE-013: PR #184621

Existing PR contributor waitwaiting on CI/check fix3 issues1 PRhigh
Issues #182649#182651#182652
PRs PR #184621
  • Diagnosis: PR #184621 is blocked by failing CI checks: statusCheckRollup, unit-test / inductor-test / test-osdc (inductor, 1, 2, mt-l-x86aavx2-29-113-a10g), inductor-test / test-osdc (inductor_timm, 2, 2, mt-l-x86aavx2-29-113-a1...
  • Next action: Fix or ask the contributor to fix statusCheckRollup, unit-test / inductor-test / test-osdc (inductor, 1, 2, mt-l-x86aavx2-29-113-a10g), inductor-test / test-osdc (inductor_timm, 2, 2, mt-l-x86aavx2-29-113-a10g), induc...
  • Evidence: failed checks: statusCheckRollup, unit-test / inductor-test / test-osdc (inductor, 1, 2, mt-l-x86aavx2-29-113-a10g), inductor-test / test-osdc (inductor_timm, 2, 2, mt-l-x86aavx2-29-113-a10g), inductor-test / test-osd...
  • Short excerpt: or scalar values. This covers tensor-backed reshape shapes from buffers and equivalent ONNX-converted reshape patterns. Fixes #182649 Fixes #182651 Fixes #182652 Generated by my agent cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzhe...
  • Review threads: thread:issue:182649:issue:182651:issue:182652:pr:184621
  • Package receipt: d4519464a69d

Bundle BUNDLE-014: PR #189294 / PR #189295

Existing PR blocked reviewwaiting on CI/check fix2 issues2 PRshigh
Issues #173059#189034
PRs PR #189294PR #189295
  • Diagnosis: PR #189294 is blocked by failing CI checks: statusCheckRollup.
  • Next action: Fix or ask the contributor to fix statusCheckRollup, rerun checks, then revisit review/owner-gate approval once the branch is green.
  • Evidence: failed checks: statusCheckRollup.
  • Short excerpt: o the numpy/scipy pair is consistent. Mirrors the compatible numpy==2.0.2 + scipy==1.13.1 set already used in test.sh's numpy_2 path. Fixes #189034. Test Plan: CI: the "Limited CI on H100 / ...sm90 / test (smoke)" job should go green; test_foreach no longer crashes importing s...
  • Review threads: thread:issue:173059:issue:189034:pr:189294:pr:189295
  • Package receipt: ac4d167facc9

Bundle BUNDLE-017: PR #185496

Existing PR contributor waitwaiting on CI/check fix2 issues1 PRhigh
Issues #146719#155044
PRs PR #185496
  • Diagnosis: PR #185496 is blocked by failing CI checks: statusCheckRollup, inductor-test / test-osdc (inductor_timm, 2, 2, mt-l-x86aavx2-29-113-a10g), inductor-test / test-osdc (inductor_timm, 1, 2, mt-l-x86aavx2-29-113-a10g), in...
  • Next action: Fix or ask the contributor to fix statusCheckRollup, inductor-test / test-osdc (inductor_timm, 2, 2, mt-l-x86aavx2-29-113-a10g), inductor-test / test-osdc (inductor_timm, 1, 2, mt-l-x86aavx2-29-113-a10g), inductor-tes...
  • Evidence: failed checks: statusCheckRollup, inductor-test / test-osdc (inductor_timm, 2, 2, mt-l-x86aavx2-29-113-a10g), inductor-test / test-osdc (inductor_timm, 1, 2, mt-l-x86aavx2-29-113-a10g), inductor-test / test-osdc (indu...
  • Short excerpt: alone, handles dynamic-shape SymInt metadata, and falls back rather than guessing if the exact alias key is ambiguous. Fixes #155044 Fixes #146719 Generated by my agent Test Plan: python test/export/test_export.py TestExport.test_autograd_grad_reuses_saved_forward_tensor TestE...
  • Review threads: thread:issue:146719:issue:155044:pr:185496
  • Package receipt: dc31c940a8f3

Bundle BUNDLE-019: PR #184628

Existing PR blocked reviewwaiting on CI/check fix2 issues1 PRhigh
Issues #182217#183369
PRs PR #184628
  • Diagnosis: PR #184628 is blocked by failing CI checks: statusCheckRollup.
  • Next action: Fix or ask the contributor to fix statusCheckRollup, rerun checks, then revisit review/owner-gate approval once the branch is green.
  • Evidence: failed checks: statusCheckRollup.
  • Short excerpt: tedTensor keepdim reductions do not create unguardable ephemeral guards. Leave mixed or unsupported SingletonInt expressions unknown. Fixes #182217 Fixes #183369 Generated by my agent cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzhe...
  • Review threads: thread:issue:182217:issue:183369:pr:184628
  • Package receipt: d9335908b10a
472 more bundle receipt(s)

Remaining bundles stay in the downloadable public-safe CSV artifacts to keep this action tab fast to scan.

fix failing static/type checks

Repair-ready PRs
Why

The failing job names are known, so maintainers can ask for or apply a focused check fix before re-review.

Next action

Fix or request fixes for the named static/type checks, rerun, then return to review only after the branch is green.

Impact

1 bundle | 0 issues | 10 PRs
Confidence: high

Evidence

failed static/type checksfailed checks: statusCheckRollup, linux-jammy-aarch64-p...1 package receipt

View bundles

Bundle BUNDLE-637: PR #176657 / PR #176670 / PR #176754 / PR #179820 / +6

Existing PR blocked reviewwaiting on CI/check fix0 issues10 PRshigh
Issues none
PRs PR #176657PR #176670PR #176754PR #179820PR #180641PR #182074PR #182216PR #188115PR #188165PR #189202
  • Diagnosis: PR #188115 is blocked by failing static/type checks: statusCheckRollup, linux-jammy-aarch64-py3.10 / build-osdc, linux-jammy-cuda13.0-py3.10-gcc11-sm90 / test-osdc (smoke, 1, 1, mt-l-x86iamx-22-225-h100), linux-jammy-...
  • Next action: Fix or ask the contributor to fix statusCheckRollup, linux-jammy-aarch64-py3.10 / build-osdc, linux-jammy-cuda13.0-py3.10-gcc11-sm90 / test-osdc (smoke, 1, 1, mt-l-x86iamx-22-225-h100), linux-jammy-cuda13.0-py3.12-gcc...
  • Evidence: failed checks: statusCheckRollup, linux-jammy-aarch64-py3.10 / build-osdc, linux-jammy-cuda13.0-py3.10-gcc11-sm90 / test-osdc (smoke, 1, 1, mt-l-x86iamx-22-225-h100), linux-jammy-cuda13.0-py3.12-gcc11-sm100 / test-osd...
  • Short excerpt: PR #188115: PR blocked; checks: statusCheckRollup, linux-jammy-aarch64-py3.10 / build-osdc, linux-jammy-cuda13.0-py3.10-gcc11-sm90 / test-osdc (smoke, 1, 1, mt-l-x86iamx-22-225-h100), linux-jammy-cuda13.0-py3.12-gcc11-sm100 / test-osdc (smoke_b200, 1, 1, mt-l-x86iamx-22-225-b2...
  • Package receipt: 8b4a8fdd91d5

resolve overlapping PR lineage

Superseded/lineage cleanup
Why

Multiple active or referenced PR attempts are connected, so maintainers should choose the current path before separate reviews diverge.

Next action

Review the current PR, older attempts, superseding links, and review guidance together; close, park, or merge in one lineage decision.

Impact

272 bundles | 329 issues | 2179 PRs
Confidence: high

Evidence

failed static/type checksfailed CI checksrequested changesno failed-check blockerreviewDecision=APPROVED272 package receipts

View bundles

Bundle BUNDLE-005: PR #181720 / PR #189256

Existing PR decisionwaiting on maintainer review3 issues2 PRsmedium
Issues #180397#181374#188970
PRs PR #181720PR #189256
  • Diagnosis: PR #189256 has no local failed-check diagnosis; it needs a maintainer merge/close/rebase decision.
  • Next action: Review changed files, linked issues, and latest maintainer comments, then choose merge, close, rebase request, or maintainer takeover.
  • Evidence: reviewDecision=APPROVED; mergeStateStatus=CLEAN; pending=; stale_days=0
  • Short excerpt: at in the current form it gives no perf benefit over regular CPU memory, it this is a foundation for copy-on-write work Fixes #181374 Fixes #188970 Authored with assistance from Claude Code. | Please note that in the current form it gives no perf benefit over regular CPU memor...
  • Review threads: thread:issue:180397:issue:181374:issue:188970:pr:181720
  • Package receipt: b3f507176f68

Bundle BUNDLE-007: PR #181726 / PR #181727 / PR #181728 / PR #183511 / +10

Existing PR blocked reviewwaiting on CI/check fix21 issues14 PRshigh
Issues #178040#183988#186348#187988#188477#188675#188704#188706#188707#188708#188709#188710+9
PRs PR #181726PR #181727PR #181728PR #183511PR #184276PR #187315PR #187318PR #187989PR #188268PR #188931+4
  • Diagnosis: PR #189149 is blocked by failing CI checks: statusCheckRollup.
  • Next action: Fix or ask the contributor to fix statusCheckRollup, rerun checks, then revisit review/owner-gate approval once the branch is green.
  • Evidence: failed checks: statusCheckRollup.
  • Short excerpt: pr #189149 declares a closing reference to issue #186348. | pr #187989 declares a closing reference to issue #187988. | pr #188931 declares a closing reference to issue #188891. | pr #184276 declares a closing reference to issue #183988. | pr #188996 declares a closing referen...
  • Review threads: thread:issue:178040:issue:183988:issue:186348:issue:187988
  • Package receipt: 9a19487204ee

Bundle BUNDLE-008: PR #184481 / PR #184694 / PR #185891

Existing PR contributor waitwaiting on CI/check fix3 issues3 PRshigh
Issues #93617#178680#188133
PRs PR #184481PR #184694PR #185891
  • Diagnosis: PR #185891 is blocked by failing CI checks: statusCheckRollup.
  • Next action: Fix or ask the contributor to fix statusCheckRollup, rerun checks, then revisit review/owner-gate approval once the branch is green.
  • Evidence: failed checks: statusCheckRollup.
  • Short excerpt: including static traces and AOTAutograd cache hits, so alias-sensitive synthetic-base graphs are not reused for incompatible inputs. Fixes #178680 Generated by my agent cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv....
  • Review threads: thread:issue:178680:issue:188133:issue:93617:pr:184481
  • Package receipt: 3d3d3ef1fb3b

Bundle BUNDLE-009: PR #185946 / PR #186375

Existing PR blocked reviewwaiting on CI/check fix3 issues2 PRshigh
Issues #16706#185240#186374
PRs PR #185946PR #186375
  • Diagnosis: PR #185946 is blocked by failing CI checks: statusCheckRollup.
  • Next action: Fix or ask the contributor to fix statusCheckRollup, rerun checks, then revisit review/owner-gate approval once the branch is green.
  • Evidence: failed checks: statusCheckRollup.
  • Short excerpt: pr #186375 declares a closing reference to issue #186374. | pr #185946 declares a closing reference to issue #185240. | ad of falling back gracefully. Similar to #185946, this PR replaces the assertion with a warning and allows control flow to continue. Fixes #186374 cc @eqy |...
  • Review threads: thread:issue:16706:issue:185240:issue:186374:pr:185946
  • Package receipt: ba5b87b237c1

Bundle BUNDLE-010: PR #185846 / PR #186358 / PR #187609

Existing PR blocked reviewwaiting on CI/check fix2 issues3 PRshigh
Issues #185470#185484
PRs PR #185846PR #186358PR #187609
  • Diagnosis: PR #186358 is blocked by failing CI checks: statusCheckRollup.
  • Next action: Fix or ask the contributor to fix statusCheckRollup, rerun checks, then revisit review/owner-gate approval once the branch is green.
  • Evidence: failed checks: statusCheckRollup.
  • Short excerpt: pr #186358 declares a closing reference to issue #185484. | so output dtype semantics are unchanged. A prior stale PR, #185846, made the same root-cause fix but has not landed on current main. Fixes #185470 Generated by my agent Test Plan: python test/inductor/test_torchinduct...
  • Review threads: thread:issue:185470:issue:185484:pr:185846:pr:186358
  • Package receipt: 221d1417621c

Bundle BUNDLE-014: PR #189294 / PR #189295

Existing PR blocked reviewwaiting on CI/check fix2 issues2 PRshigh
Issues #173059#189034
PRs PR #189294PR #189295
  • Diagnosis: PR #189294 is blocked by failing CI checks: statusCheckRollup.
  • Next action: Fix or ask the contributor to fix statusCheckRollup, rerun checks, then revisit review/owner-gate approval once the branch is green.
  • Evidence: failed checks: statusCheckRollup.
  • Short excerpt: o the numpy/scipy pair is consistent. Mirrors the compatible numpy==2.0.2 + scipy==1.13.1 set already used in test.sh's numpy_2 path. Fixes #189034. Test Plan: CI: the "Limited CI on H100 / ...sm90 / test (smoke)" job should go green; test_foreach no longer crashes importing s...
  • Review threads: thread:issue:173059:issue:189034:pr:189294:pr:189295
  • Package receipt: ac4d167facc9

Bundle BUNDLE-015: PR #187354 / PR #188556

Existing PR decisionwaiting on maintainer review2 issues2 PRsmedium
Issues #187332#188545
PRs PR #187354PR #188556
  • Diagnosis: PR #187354 has no local failed-check diagnosis; it needs a maintainer merge/close/rebase decision.
  • Next action: Review changed files, linked issues, and latest maintainer comments, then choose merge, close, rebase request, or maintainer takeover.
  • Evidence: reviewDecision=APPROVED; mergeStateStatus=CLEAN; pending=; stale_days=0
  • Short excerpt: pr #188556 declares a closing reference to issue #188545. | Fixes #188545 Summary Preserve eager-mode semantics for infinite inputs in the Triton lowering of torch.special.bessel_j0 torch.special.bessel_j1 torch.s | ile continuing to use libdevice for finite inputs. This prese...
  • Review threads: thread:issue:187332:issue:188545:pr:187354:pr:188556
  • Package receipt: 0c1a21954532

Bundle BUNDLE-016: PR #185756 / PR #187908

Existing PR contributor waitwaiting on contributor2 issues2 PRsmedium
Issues #171356#187429
PRs PR #185756PR #187908
  • Diagnosis: PR #187908 is contributor-owned until requested changes or review questions are answered.
  • Next action: Keep parked or close after maintainer review if the contributor path is no longer active; do not ask for maintainer work first.
  • Evidence: reviewDecision=CHANGES_REQUESTED; stale_days=0
  • Short excerpt: pr #187908 declares a closing reference to issue #187429. | pr #185756 declares a closing reference to issue #171356. | Fixes #187429 Root cause Commit 0eae6b68f42 ("Unify torch.tensor and torch.ops.aten.scalar_tensor behavior") intentionally relaxed the cast in fill_inpla | t...
  • Review threads: thread:issue:171356:issue:187429:pr:185756:pr:187908
  • Package receipt: a8943e8c850c

Bundle BUNDLE-082: PR #183884 / PR #184161 / PR #185433 / PR #185976 / +3

Existing PR contributor waitwaiting on CI/check fix10 issues7 PRshigh
Issues #130182#136216#137520#140842#144360#145093#145231#158088#186031#186136
PRs PR #183884PR #184161PR #185433PR #185976PR #186011PR #186094PR #186361
  • Diagnosis: PR #186361 is blocked by failing CI checks: statusCheckRollup, torchtitan-x-pytorch-test / test-osdc (torchtitan_features_integration, 1, 1, mt-l-x86aavx2-189-704-a10g-8), inductor-test / test-osdc (inductor_torchbenc...
  • Next action: Fix or ask the contributor to fix statusCheckRollup, torchtitan-x-pytorch-test / test-osdc (torchtitan_features_integration, 1, 1, mt-l-x86aavx2-189-704-a10g-8), inductor-test / test-osdc (inductor_torchbench, 2, 2, m...
  • Evidence: failed checks: statusCheckRollup, torchtitan-x-pytorch-test / test-osdc (torchtitan_features_integration, 1, 1, mt-l-x86aavx2-189-704-a10g-8), inductor-test / test-osdc (inductor_torchbench, 2, 2, mt-l-x86aavx2-29-113...
  • Short excerpt: g only actually used unbacked symbol definitions, instead of pinning every node that defines an unbacked symbol in DCE. Fixes #140842 Fixes #137520 Generated by my agent Benchmark Results: A small Python microbenchmark compared the previous scanner predicate (SymTypes, Expr) w...
  • Review threads: thread:issue:130182:issue:136216:issue:137520:issue:140842
  • Package receipt: afdc55cff25b

Bundle BUNDLE-083: PR #184975 / PR #184976 / PR #185953 / PR #186801 / +1

Existing PR decisionwaiting on CI/check fix7 issues5 PRshigh
Issues #145529#153056#170137#170770#170781#186796#187008
PRs PR #184975PR #184976PR #185953PR #186801PR #187090
  • Diagnosis: PR #185953 is blocked by failing CI checks: statusCheckRollup, torchtitan-x-pytorch-test / test-osdc (torchtitan_features_integration, 1, 1, lf-l-x86aavx2-189-704-a10g-8), inductor-test / test-osdc (inductor_timm, 2,...
  • Next action: Fix or ask the contributor to fix statusCheckRollup, torchtitan-x-pytorch-test / test-osdc (torchtitan_features_integration, 1, 1, lf-l-x86aavx2-189-704-a10g-8), inductor-test / test-osdc (inductor_timm, 2, 2, mt-l-x8...
  • Evidence: failed checks: statusCheckRollup, torchtitan-x-pytorch-test / test-osdc (torchtitan_features_integration, 1, 1, lf-l-x86aavx2-189-704-a10g-8), inductor-test / test-osdc (inductor_timm, 2, 2, mt-l-x86aavx2-29-113-a10g)...
  • Short excerpt: n-TensorWeakRef weakrefs. Ordinary user weakrefs remain blocked because they can observe a tensor object whose contents were swapped. Fixes #186796 Generated by my agent Test Plan: python test/test_torch.py -k test_swap_allows_tensor_weakref python test/test_torch.py TestTorch...
  • Review threads: thread:issue:145529:issue:153056:issue:170137:issue:170770
  • Package receipt: 6a66b47c844f

Bundle BUNDLE-084: PR #184978 / PR #185059

Existing PR decisionwaiting on maintainer review3 issues2 PRsmedium
Issues #167007#170672#188153
PRs PR #184978PR #185059
  • Diagnosis: PR #185059 has no local failed-check diagnosis; it needs a maintainer merge/close/rebase decision.
  • Next action: Review changed files, linked issues, and latest maintainer comments, then choose merge, close, rebase request, or maintainer takeover.
  • Evidence: reviewDecision=; mergeStateStatus=CLEAN; pending=; stale_days=0
  • Short excerpt: xport sees a bogus lifted fake tensor constant, it used to raise an internal diagnostic ending with "Please file an issue on github." Issue #170672 hits this path by reading parameters through .data before copying back into the parameter under no_grad. Direct no_grad parameter...
  • Review threads: thread:issue:167007:issue:170672:issue:188153:pr:184978
  • Package receipt: 85bf234b0f57

Bundle BUNDLE-088: PR #185170 / PR #186418

Existing PR decisionwaiting on maintainer review2 issues2 PRsmedium
Issues #128554#161671
PRs PR #185170PR #186418
  • Diagnosis: PR #186418 has no local failed-check diagnosis; it needs a maintainer merge/close/rebase decision.
  • Next action: Review changed files, linked issues, and latest maintainer comments, then choose merge, close, rebase request, or maintainer takeover.
  • Evidence: reviewDecision=; mergeStateStatus=CLEAN; pending=; stale_days=0
  • Short excerpt: izing the entire GraphModule as an opaque nested byte blob, which would hide its tensor storages from the outer torch.save machinery. Fixes #161671 Generated by my agent Test Plan: python test/export/test_serialize.py TestSaveLoad.test_torch_save_exported_program_with_call_mod...
  • Review threads: thread:issue:128554:issue:161671:pr:185170:pr:186418
  • Package receipt: 9194927cc301
260 more bundle receipt(s)

Remaining bundles stay in the downloadable public-safe CSV artifacts to keep this action tab fast to scan.

park or close contributor-waiting PRs

Batch issue sweeps
Why

The next work is contributor-owned unless a maintainer chooses to take over or close the stale path.

Next action

Keep parked, close with context, or explicitly take over only when the evidence warrants maintainer work.

Impact

46 bundles | 48 issues | 46 PRs
Confidence: medium

Evidence

requested changesreviewDecision=CHANGES_REQUESTEDstale_days=6stale_days=046 package receipts

View bundles

Bundle BUNDLE-001: PR #188801

Existing PR contributor waitwaiting on contributor1 issue1 PRmedium
Issues #188812
PRs PR #188801
  • Diagnosis: PR #188801 is contributor-owned until requested changes or review questions are answered.
  • Next action: Keep parked or close after maintainer review if the contributor path is no longer active; do not ask for maintainer work first.
  • Evidence: reviewDecision=CHANGES_REQUESTED; stale_days=6
  • Short excerpt: MetalShaderLibrary& MetalShaderLibrary::getBundledLibrary() { static BundledShaderLibrary* l = new BundledShaderLibrary(); return *l; } PR: #188801 (will be scoped down to just this leak). cc @malfet @aditvenk @kulinseth @DenisVieriu97 @jhavukainen @Isalia20 Versions PyTorch v...
  • Review threads: thread:issue:188812:pr:188801
  • Package receipt: 33311143c20a

Bundle BUNDLE-018: PR #184632

Existing PR contributor waitwaiting on contributor2 issues1 PRmedium
Issues #182200#184101
PRs PR #184632
  • Diagnosis: PR #184632 is contributor-owned until requested changes or review questions are answered.
  • Next action: Keep parked or close after maintainer review if the contributor path is no longer active; do not ask for maintainer work first.
  • Evidence: reviewDecision=CHANGES_REQUESTED; stale_days=0
  • Short excerpt: ch path and the fast binary path, and covers the real-valued promotion ops including copysign and transposed complex tensor variants. Fixes #182200 Fixes #184101 Generated by my agent Test Plan: lintrunner -a python test/test_meta.py -k test_cuda_mixed_dtype_binary_pointwise_f...
  • Review threads: thread:issue:182200:issue:184101:pr:184632
  • Package receipt: 45493fed498f

Bundle BUNDLE-026: PR #188605

Existing PR contributor waitwaiting on contributor1 issue1 PRmedium
Issues #188547
PRs PR #188605
  • Diagnosis: PR #188605 is contributor-owned until requested changes or review questions are answered.
  • Next action: Keep parked or close after maintainer review if the contributor path is no longer active; do not ask for maintainer work first.
  • Evidence: reviewDecision=CHANGES_REQUESTED; stale_days=1
  • Short excerpt: pr #188605 declares a closing reference to issue #188547. | Fixes #188547 Fixes infinite recursion when calling int( ), float( ), or indexing pybind11 enums in compiled functions. Basically detects GetAttrVariable | Overall looks good, I think we can simplify the test, thanks!...
  • Review threads: thread:issue:188547:pr:188605
  • Package receipt: 85d8b9894232

Bundle BUNDLE-031: PR #187955

Existing PR contributor waitwaiting on contributor1 issue1 PRmedium
Issues #188008
PRs PR #187955
  • Diagnosis: PR #187955 is contributor-owned until requested changes or review questions are answered.
  • Next action: Keep parked or close after maintainer review if the contributor path is no longer active; do not ask for maintainer work first.
  • Evidence: reviewDecision=CHANGES_REQUESTED; stale_days=10
  • Short excerpt: Being fixed in #187955: mapAndSetAccess is now strongly exception-safe — on a partial map/setAccess failure it unmaps the mapped prefix and releases the range's h | ing unwinding → std::terminate. This is a pre-existing latent bug (the per-segment unmap loop had the same defec...
  • Review threads: thread:issue:188008:pr:187955
  • Package receipt: 8804b9ddbffb

Bundle BUNDLE-042: PR #184970

Existing PR contributor waitwaiting on contributor1 issue1 PRmedium
Issues #171894
PRs PR #184970
  • Diagnosis: PR #184970 is contributor-owned until requested changes or review questions are answered.
  • Next action: Keep parked or close after maintainer review if the contributor path is no longer active; do not ask for maintainer work first.
  • Evidence: reviewDecision=CHANGES_REQUESTED; stale_days=1
  • Short excerpt: I checked the abandoned #173517 attempt and kept the same root cause direction without adding a hard-coded lift_fresh_copy exception. Fixes #171894 Generated by my agent Test Plan: ninja -C build lib/libtorch_cpu.so python test/test_functionalization.py TestFunctionalization.t...
  • Review threads: thread:issue:171894:pr:184970
  • Package receipt: 6455c21e9526

Bundle BUNDLE-044: PR #184153

Existing PR contributor waitwaiting on contributor1 issue1 PRmedium
Issues #146569
PRs PR #184153
  • Diagnosis: PR #184153 is contributor-owned until requested changes or review questions are answered.
  • Next action: Keep parked or close after maintainer review if the contributor path is no longer active; do not ask for maintainer work first.
  • Evidence: reviewDecision=CHANGES_REQUESTED; stale_days=1
  • Short excerpt: nvalidate live .grad storage. Clone backward outputs before handing them to autograd so persistent grads have normal tensor lifetime. Fixes #146569 Generated by my agent cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv...
  • Review threads: thread:issue:146569:pr:184153
  • Package receipt: 062607425baa

Bundle BUNDLE-068: PR #184961

Existing PR contributor waitwaiting on contributor1 issue1 PRmedium
Issues #171978
PRs PR #184961
  • Diagnosis: PR #184961 is contributor-owned until requested changes or review questions are answered.
  • Next action: Keep parked or close after maintainer review if the contributor path is no longer active; do not ask for maintainer work first.
  • Evidence: reviewDecision=CHANGES_REQUESTED; stale_days=1
  • Short excerpt: nd would still allow wrong shapes for max(lengths) < padded_length, so this patch handles the packed-sequence metadata path directly. Fixes #171978 Generated by my agent Test Plan: python test/export/test_export.py -k test_export_packed_sequence python test/export/test_export....
  • Review threads: thread:issue:171978:pr:184961
  • Package receipt: 4db135ae2adf

Bundle BUNDLE-087: PR #185124

Existing PR contributor waitwaiting on contributor3 issues1 PRmedium
Issues #157610#162734#163037
PRs PR #185124
  • Diagnosis: PR #185124 is contributor-owned until requested changes or review questions are answered.
  • Next action: Keep parked or close after maintainer review if the contributor path is no longer active; do not ask for maintainer work first.
  • Evidence: reviewDecision=CHANGES_REQUESTED; stale_days=1
  • Short excerpt: s run_decompositions during export — all of which previously failed with the same alias-annotation error. Fixes #163037 Fixes #162734 Fixes #157610 Generated by my agent Test Plan: python setup.py develop python - <<'PY' ... torch.compile(ReproBroadcastInDim(), backend="induct...
  • Review threads: thread:issue:157610:issue:162734:issue:163037:pr:185124
  • Package receipt: dd2feaee0bd2

Bundle BUNDLE-104: PR #185499

Existing PR contributor waitwaiting on contributor1 issue1 PRmedium
Issues #154647
PRs PR #185499
  • Diagnosis: PR #185499 is contributor-owned until requested changes or review questions are answered.
  • Next action: Keep parked or close after maintainer review if the contributor path is no longer active; do not ask for maintainer work first.
  • Evidence: reviewDecision=CHANGES_REQUESTED; stale_days=1
  • Short excerpt: d not remove nonzero_memo or skip the assertion because those would hide the missing fresh-binding invariant rather than maintain it. Fixes #154647 Generated by my agent Test Plan: Issue repro before fix: failed in ep.run_decompositions(...) with AssertionError: u2 -> u0 from....
  • Review threads: thread:issue:154647:pr:185499
  • Package receipt: ad53a5e80357

Bundle BUNDLE-109: PR #185353

Existing PR contributor waitwaiting on contributor1 issue1 PRmedium
Issues #159918
PRs PR #185353
  • Diagnosis: PR #185353 is contributor-owned until requested changes or review questions are answered.
  • Next action: Keep parked or close after maintainer review if the contributor path is no longer active; do not ask for maintainer work first.
  • Evidence: reviewDecision=CHANGES_REQUESTED; stale_days=1
  • Short excerpt: original1 naming globally, since those names are part of parametrization internals and changing them would be a much broader BC risk. Fixes #159918 Generated by my agent Test Plan: python test/export/test_export.py TestExport.test_subclasses_parameterization TestExport.test_su...
  • Review threads: thread:issue:159918:pr:185353
  • Package receipt: 6d0331b30ff8

Bundle BUNDLE-110: PR #185150

Existing PR contributor waitwaiting on contributor1 issue1 PRmedium
Issues #161807
PRs PR #185150
  • Diagnosis: PR #185150 is contributor-owned until requested changes or review questions are answered.
  • Next action: Keep parked or close after maintainer review if the contributor path is no longer active; do not ask for maintainer work first.
  • Evidence: reviewDecision=CHANGES_REQUESTED; stale_days=0
  • Short excerpt: ules, but that would only hide the source bug: the GraphModule produced by FX tracing was already returning the wrong container type. Fixes #161807 Generated by my agent Test Plan: python test/test_fx.py TestFX.test_symbolic_trace_preserves_ordered_dict_output (blocked before....
  • Review threads: thread:issue:161807:pr:185150
  • Package receipt: 3512fb8924db

Bundle BUNDLE-161: PR #185043

Existing PR contributor waitwaiting on contributor2 issues1 PRmedium
Issues #167568#182403
PRs PR #185043
  • Diagnosis: PR #185043 is contributor-owned until requested changes or review questions are answered.
  • Next action: Keep parked or close after maintainer review if the contributor path is no longer active; do not ask for maintainer work first.
  • Evidence: reviewDecision=CHANGES_REQUESTED; stale_days=1
  • Short excerpt: ntains triton_kernel_wrapper_functional, no longer contains the mutating wrapper, and returns the functional wrapper's cloned output. Fixes #167568 Generated by my agent Test Plan: python test/higher_order_ops/test_local_map.py TestLocalMap.test_mutations_triton_deferred_inlin...
  • Review threads: thread:issue:167568:issue:182403:pr:185043
  • Package receipt: 73e26b1c2975
34 more bundle receipt(s)

Remaining bundles stay in the downloadable public-safe CSV artifacts to keep this action tab fast to scan.

consolidate duplicate candidates

Batch issue sweeps
Why

Related reports can share one canonical answer, close path, or implementation direction.

Next action

Pick the canonical issue, preserve the strongest reproduction, and close or link duplicates in one pass.

Impact

22 bundles | 184 issues | 0 PRs
Confidence: medium

Evidence

issues #167693 and #119536 share title/component terms:...issues #167693 and #119346 share title/component terms:...issues #167693 and #119364 share title/component terms:...issues #167693 and #119357 share title/component terms:...22 package receipts

View bundles

Bundle BUNDLE-864: module: __torch_function__ needs triage queue (24 issues)

Duplicate/related issue sweepneeds design decision24 issues0 PRsmedium
Issues #119346#119357#119364#119490#119514#119536#119537#119576#167693#185267#185268#186344+12
PRs none
  • Diagnosis: The local evidence indicates likely related reports that should share one canonical answer or implementation path.
  • Next action: Pick the canonical issue, preserve the strongest reproduction, and close/link duplicates in one maintainer pass.
  • Evidence: issues #167693 and #119536 share title/component terms: __main__, disabled | issues #167693 and #119346 share title/component terms: __main__, disabled | issues #167693 and #119364 share title/component terms: __main_...
  • Short excerpt: issues #167693 and #119536 share title/component terms: __main__, disabled | issues #167693 and #119346 share title/component terms: __main__, disabled | issues #167693 and #119364 share title/component terms: __main__, disabled | issues #167693 and #119357 share title/compone...
  • Package receipt: 4c678752690c

Bundle BUNDLE-865: module: abi high-priority/blocker queue (2 issues)

Duplicate/related issue sweephigh-priority/blocker2 issues0 PRsmedium
Issues #149658#187320
PRs none
  • Diagnosis: The local evidence indicates likely related reports that should share one canonical answer or implementation path.
  • Next action: Pick the canonical issue, preserve the strongest reproduction, and close/link duplicates in one maintainer pass.
  • Evidence: issues #187320 and #149658 share title/component terms: pytorch, stable, version
  • Short excerpt: issues #187320 and #149658 share title/component terms: pytorch, stable, version
  • Package receipt: 1971cae443f5

Bundle BUNDLE-866: module: activation checkpointing needs triage queue (2 issues)

Duplicate/related issue sweepneeds design decision2 issues0 PRsmedium
Issues #85088#183207
PRs none
  • Diagnosis: The local evidence indicates likely related reports that should share one canonical answer or implementation path.
  • Next action: Pick the canonical issue, preserve the strongest reproduction, and close/link duplicates in one maintainer pass.
  • Evidence: issues #85088 and #183207 share title/component terms: does, not, work
  • Short excerpt: issues #85088 and #183207 share title/component terms: does, not, work
  • Package receipt: 07b9787d7c5d

Bundle BUNDLE-867: module: amp (automated mixed precision) high-priority/blocker queue (49 issues)

Duplicate/related issue sweephigh-priority/blocker52 issues0 PRsmedium
Issues #30651#55669#66782#78829#91633#96779#97086#111901#116423#117494#118088#119845+40
PRs none
  • Diagnosis: The local evidence indicates likely related reports that should share one canonical answer or implementation path.
  • Next action: Pick the canonical issue, preserve the strongest reproduction, and close/link duplicates in one maintainer pass.
  • Evidence: issues #116423 and #155146 share title/component terms: fault, segmentation | issues #116423 and #183251 share title/component terms: fault, pytorch, segmentation | issues #119845 and #155146 share title/component ter...
  • Short excerpt: issues #116423 and #155146 share title/component terms: fault, segmentation | issues #116423 and #183251 share title/component terms: fault, pytorch, segmentation | issues #119845 and #155146 share title/component terms: fault, segmentation | issues #119845 and #181853 share t...
  • Package receipt: 46341333cbf5

Bundle BUNDLE-868: module: aotdispatch high-priority/blocker queue (6 issues)

Duplicate/related issue sweephigh-priority/blocker6 issues0 PRsmedium
Issues #38910#173313#174468#174704#183890#187781
PRs none
  • Diagnosis: The local evidence indicates likely related reports that should share one canonical answer or implementation path.
  • Next action: Pick the canonical issue, preserve the strongest reproduction, and close/link duplicates in one maintainer pass.
  • Evidence: issues #174468 and #174704 share title/component terms: compile, torch | issues #174468 and #173313 share title/component terms: compile, torch | issues #174468 and #183890 share title/component terms: compile, torch...
  • Short excerpt: issues #174468 and #174704 share title/component terms: compile, torch | issues #174468 and #173313 share title/component terms: compile, torch | issues #174468 and #183890 share title/component terms: compile, torch | issues #174468 and #187781 share title/component terms: co...
  • Package receipt: 85fe06db2c61

Bundle BUNDLE-869: module: arm has linked PR queue (2 issues)

Duplicate/related issue sweephas linked PR9 issues0 PRsmedium
Issues #40425#149201#161783#162422#170877#177247#177249#178538#188255
PRs none
  • Diagnosis: The local evidence indicates likely related reports that should share one canonical answer or implementation path.
  • Next action: Pick the canonical issue, preserve the strongest reproduction, and close/link duplicates in one maintainer pass.
  • Evidence: issues #162422 and #161783 share title/component terms: compile, runtime, torch | issues #188255 and #178538 share title/component terms: crash, import, torch | issues #170877 and #177249 share title/component terms:...
  • Short excerpt: issues #162422 and #161783 share title/component terms: compile, runtime, torch | issues #188255 and #178538 share title/component terms: crash, import, torch | issues #170877 and #177249 share title/component terms: aarch64, failure, test, unit | issues #170877 and #177247 sh...
  • Package receipt: 87c1606ebbd1

Bundle BUNDLE-870: module: autograd needs triage queue (4 issues)

Duplicate/related issue sweepneeds design decision16 issues0 PRsmedium
Issues #73294#80946#143580#146907#153937#155693#158701#172798#181024#182283#185715#186453+4
PRs none
  • Diagnosis: The local evidence indicates likely related reports that should share one canonical answer or implementation path.
  • Next action: Pick the canonical issue, preserve the strongest reproduction, and close/link duplicates in one maintainer pass.
  • Evidence: issues #189282 and #189281 share title/component terms: 2025, aarch64, cpu, measurement, nov, operator_benchmark, possible, since | issues #189281 and #189283 share title/component terms: 2025, cpu, nov, operator_benc...
  • Short excerpt: issues #189282 and #189281 share title/component terms: 2025, aarch64, cpu, measurement, nov, operator_benchmark, possible, since | issues #189281 and #189283 share title/component terms: 2025, cpu, nov, operator_benchmark, since | issues #189281 and #189280 share title/compon...
  • Package receipt: 837d247a835c

Bundle BUNDLE-871: module: bfloat16 needs triage queue (14 issues)

Duplicate/related issue sweepneeds design decision14 issues0 PRsmedium
Issues #19106#48251#52741#67797#75171#86532#86594#86595#97876#159238#173492#183812+2
PRs none
  • Diagnosis: The local evidence indicates likely related reports that should share one canonical answer or implementation path.
  • Next action: Pick the canonical issue, preserve the strongest reproduction, and close/link duplicates in one maintainer pass.
  • Evidence: issues #159238 and #19106 share title/component terms: jit, performance, torch, trace | issues #86595 and #86594 share title/component terms: computing, crash, jit, model, will | issues #86595 and #86532 share title/c...
  • Short excerpt: issues #159238 and #19106 share title/component terms: jit, performance, torch, trace | issues #86595 and #86594 share title/component terms: computing, crash, jit, model, will | issues #86595 and #86532 share title/component terms: crash, jit, will | issues #86594 and #86532...
  • Package receipt: 65c075d77ef5

Bundle BUNDLE-872: module: binaries high-priority/blocker queue (2 issues)

Duplicate/related issue sweephigh-priority/blocker8 issues0 PRsmedium
Issues #37449#66033#87209#119547#170039#178085#188958#188959
PRs none
  • Diagnosis: The local evidence indicates likely related reports that should share one canonical answer or implementation path.
  • Next action: Pick the canonical issue, preserve the strongest reproduction, and close/link duplicates in one maintainer pass.
  • Evidence: issues #170039 and #66033 share title/component terms: __1, argument, failed, invalid, std, system_error | issues #178085 and #87209 share title/component terms: access, cuda, encountered, error, illegal, memory, was...
  • Short excerpt: issues #170039 and #66033 share title/component terms: __1, argument, failed, invalid, std, system_error | issues #178085 and #87209 share title/component terms: access, cuda, encountered, error, illegal, memory, was | issues #178085 and #37449 share title/component terms: acc...
  • Package receipt: 1cc78b9d40ca

Bundle BUNDLE-873: module: c10d needs triage queue (3 issues)

Duplicate/related issue sweepneeds design decision3 issues0 PRsmedium
Issues #74809#144359#163193
PRs none
  • Diagnosis: The local evidence indicates likely related reports that should share one canonical answer or implementation path.
  • Next action: Pick the canonical issue, preserve the strongest reproduction, and close/link duplicates in one maintainer pass.
  • Evidence: issues #163193 and #144359 share title/component terms: incorrect, results | issues #144359 and #74809 share title/component terms: incorrect, results, tensor
  • Short excerpt: issues #163193 and #144359 share title/component terms: incorrect, results | issues #144359 and #74809 share title/component terms: incorrect, results, tensor
  • Package receipt: ef65832281c6

Bundle BUNDLE-874: module: complex high-priority/blocker queue (2 issues)

Duplicate/related issue sweephigh-priority/blocker2 issues0 PRsmedium
Issues #185405#188069
PRs none
  • Diagnosis: The local evidence indicates likely related reports that should share one canonical answer or implementation path.
  • Next action: Pick the canonical issue, preserve the strongest reproduction, and close/link duplicates in one maintainer pass.
  • Evidence: issues #188069 and #185405 share title/component terms: autotune, bmm, compile, max, mode, torch
  • Short excerpt: issues #188069 and #185405 share title/component terms: autotune, bmm, compile, max, mode, torch
  • Package receipt: f29ebe5995bb

Bundle BUNDLE-875: module: correctness (silent) high-priority/blocker queue (6 issues)

Duplicate/related issue sweephigh-priority/blocker6 issues0 PRsmedium
Issues #187444#187445#187446#187447#187448#187449
PRs none
  • Diagnosis: The local evidence indicates likely related reports that should share one canonical answer or implementation path.
  • Next action: Pick the canonical issue, preserve the strongest reproduction, and close/link duplicates in one maintainer pass.
  • Evidence: issues #187444 and #187449 share title/component terms: adamw, compile, ddp, divergence, eager, nccl, node, resnet50 | issues #187444 and #187448 share title/component terms: adamw, compile, ddp, divergence, eager, re...
  • Short excerpt: issues #187444 and #187449 share title/component terms: adamw, compile, ddp, divergence, eager, nccl, node, resnet50 | issues #187444 and #187448 share title/component terms: adamw, compile, ddp, divergence, eager, resnet50, torch | issues #187444 and #187447 share title/compo...
  • Package receipt: 160852e585ce
10 more bundle receipt(s)

Remaining bundles stay in the downloadable public-safe CSV artifacts to keep this action tab fast to scan.

apply stale sweep policy

Batch issue sweeps
Why

Old low-urgency rows can be handled as a policy batch instead of reopened one at a time.

Next action

Close with rationale when no live reproducer exists, mark canonical duplicates, or promote still-reproducible failures into a local draft.

Impact

281 bundles | 9771 issues | 0 PRs
Confidence: medium

Evidence

issue #103382: stale/low urgencylabels: triagedtitle: NotImplementedError Could not run 'c10d::alltoal...issue #102921: stale/low urgency281 package receipts

View bundles

Bundle BUNDLE-1177: general stale/low urgency batch 9

Stale issue resolution sweepstale/low urgency140 issues0 PRsmedium
Issues #4073#6552#6846#6901#7047#7227#7480#7493#7571#7729#7781#7789+128
PRs none
  • Diagnosis: This is a general stale issue batch, not a single code diagnosis.
  • Next action: Apply one sweep policy: close with rationale when no live reproducer exists, mark canonical duplicates, or promote still-reproducible failures into a local patch-plan bundle.
  • Evidence: issue #103382: stale/low urgency; labels: triaged; title: NotImplementedError Could not run 'c10d::alltoall_' with arguments from the 'Meta' backend. | issue #102921: stale/low urgency; labels: triaged; title: Unbox e...
  • Short excerpt: issue #103382: stale/low urgency; labels: triaged; title: NotImplementedError Could not run 'c10d::alltoall_' with arguments from the 'Meta' backend. | issue #102921: stale/low urgency; labels: triaged; title: Unbox expectedml | issue #102741: stale/low urgency; labels: triage...
  • Package receipt: 9a1679f2227a

Bundle BUNDLE-1178: general stale/low urgency batch 41

Stale issue resolution sweepstale/low urgency140 issues0 PRsmedium
Issues #6618#6660#6868#7078#7912#8471#8937#8944#9407#9531#9676#9704+128
PRs none
  • Diagnosis: This is a general stale issue batch, not a single code diagnosis.
  • Next action: Apply one sweep policy: close with rationale when no live reproducer exists, mark canonical duplicates, or promote still-reproducible failures into a local patch-plan bundle.
  • Evidence: issue #11890: stale/low urgency; labels: caffe2; title: caffe2 argmax and argmin documentation incorrect for output type | issue #11885: stale/low urgency; labels: caffe2; title: [caffe2] Bug for softmaxwithloss opera...
  • Short excerpt: issue #11890: stale/low urgency; labels: caffe2; title: caffe2 argmax and argmin documentation incorrect for output type | issue #11885: stale/low urgency; labels: caffe2; title: [caffe2] Bug for softmaxwithloss operator | issue #9975: stale/low urgency; labels: caffe2; title:...
  • Package receipt: 223418e87611

Bundle BUNDLE-1179: general stale/low urgency batch 32

Stale issue resolution sweepstale/low urgency140 issues0 PRsmedium
Issues #2739#3176#5266#6402#6408#6613#6760#6794#8028#8550#8886#9029+128
PRs none
  • Diagnosis: This is a general stale issue batch, not a single code diagnosis.
  • Next action: Apply one sweep policy: close with rationale when no live reproducer exists, mark canonical duplicates, or promote still-reproducible failures into a local patch-plan bundle.
  • Evidence: issue #6408: stale/low urgency; labels: caffe2; title: [caffe2] Prefetching blobs for memory optimization | issue #18944: stale/low urgency; labels: caffe2; title: Caffe2 on google colab: | issue #15120: stale/low urg...
  • Short excerpt: issue #6408: stale/low urgency; labels: caffe2; title: [caffe2] Prefetching blobs for memory optimization | issue #18944: stale/low urgency; labels: caffe2; title: Caffe2 on google colab: | issue #15120: stale/low urgency; labels: caffe2; title: Different implementations of up...
  • Package receipt: 243640805cd4

Bundle BUNDLE-1180: general stale/low urgency batch 23

Stale issue resolution sweepstale/low urgency140 issues0 PRsmedium
Issues #3823#6505#7020#7546#7745#9133#9375#10913#11516#11537#12690#13034+128
PRs none
  • Diagnosis: This is a general stale issue batch, not a single code diagnosis.
  • Next action: Apply one sweep policy: close with rationale when no live reproducer exists, mark canonical duplicates, or promote still-reproducible failures into a local patch-plan bundle.
  • Evidence: issue #42173: stale/low urgency; labels: triaged, function request; title: Exclusive flag in torch.cumsum | issue #7546: stale/low urgency; labels: caffe2; title: [Caffe2] Build broken on macOS High Sierra: can't find...
  • Short excerpt: issue #42173: stale/low urgency; labels: triaged, function request; title: Exclusive flag in torch.cumsum | issue #7546: stale/low urgency; labels: caffe2; title: [Caffe2] Build broken on macOS High Sierra: can't find sys headers in /usr/local/include. | issue #3823: stale/low...
  • Package receipt: 3bb849d9ca0b

Bundle BUNDLE-1181: general stale/low urgency batch 14

Stale issue resolution sweepstale/low urgency140 issues0 PRsmedium
Issues #630#711#1442#1641#1642#1861#2569#5385#7250#9417#9624#9898+128
PRs none
  • Diagnosis: This is a general stale issue batch, not a single code diagnosis.
  • Next action: Apply one sweep policy: close with rationale when no live reproducer exists, mark canonical duplicates, or promote still-reproducible failures into a local patch-plan bundle.
  • Evidence: issue #79050: stale/low urgency; labels: triaged, lazy; title: [MetaIssue] Investigate if we should be reusing primtorch formulas for `is_dynamic` | issue #79031: stale/low urgency; labels: triaged, lazy; title: Imple...
  • Short excerpt: issue #79050: stale/low urgency; labels: triaged, lazy; title: [MetaIssue] Investigate if we should be reusing primtorch formulas for `is_dynamic` | issue #79031: stale/low urgency; labels: triaged, lazy; title: Implement SymbolicIntNode interface for lazy (i.e. lazy::Symbolic...
  • Package receipt: fbd2cf4ac40e

Bundle BUNDLE-1182: general stale/low urgency batch 47

Stale issue resolution sweepstale/low urgency33 issues0 PRsmedium
Issues #1487#6273#6357#6542#6783#6845#6857#6875#6902#7127#7133#7179+21
PRs none
  • Diagnosis: This is a general stale issue batch, not a single code diagnosis.
  • Next action: Apply one sweep policy: close with rationale when no live reproducer exists, mark canonical duplicates, or promote still-reproducible failures into a local patch-plan bundle.
  • Evidence: issue #7133: stale/low urgency; labels: caffe2; title: [Caffe2 warpctc] How to use the offered warpctc? | issue #7179: stale/low urgency; labels: caffe2; title: [caffe2]when i do CMAKE | issue #6902: stale/low urgency...
  • Short excerpt: issue #7133: stale/low urgency; labels: caffe2; title: [Caffe2 warpctc] How to use the offered warpctc? | issue #7179: stale/low urgency; labels: caffe2; title: [caffe2]when i do CMAKE | issue #6902: stale/low urgency; labels: caffe2; title: [Caffe2] Flatten Layer in caffe2 |...
  • Review threads: thread:issue:10007:issue:10254:issue:9853:issue:9854, thread:issue:68287:issue:68288:issue:68298, thread:issue:12983:issue:9484:issue:9604, +5
  • Package receipt: 2c3c5c892d41

Bundle BUNDLE-1183: Review thread: 2 issues - Issue #10119 - [Caffe2] MNIST Tutorial LMDB Error "Cannot open db"

Stale issue resolution sweepstale/low urgency2 issues0 PRsmedium
Issues #7807#10119
PRs none
  • Diagnosis: This is a general stale issue batch, not a single code diagnosis.
  • Next action: Apply one sweep policy: close with rationale when no live reproducer exists, mark canonical duplicates, or promote still-reproducible failures into a local patch-plan bundle.
  • Evidence: put - none of these combos work. I've seen other posts on this issue such as: facebookarchive/caffe2#1074 facebookarchive/caffe2#2282 #9736 #7807 facebookarchive/tutorials#7 But most of these issues were either older,...
  • Short excerpt: put - none of these combos work. I've seen other posts on this issue such as: facebookarchive/caffe2#1074 facebookarchive/caffe2#2282 #9736 #7807 facebookarchive/tutorials#7 But most of these issues were either older, posted by those who built from source or had different envi...
  • Review threads: thread:issue:10119:issue:7807
  • Package receipt: b105b63cb955

Bundle BUNDLE-1184: Review thread: 2 issues - Issue #129118 - interpolate nearest get values zero when outputs over 4G elements

Stale issue resolution sweepstale/low urgency18 issues0 PRsmedium
Issues #17897#25747#31708#84520#95024#117736#129118#133751#137368#141831#144824#147118+6
PRs none
  • Diagnosis: This is a module: 64-bit stale issue batch, not a single code diagnosis.
  • Next action: Apply one sweep policy: close with rationale when no live reproducer exists, mark canonical duplicates, or promote still-reproducible failures into a local patch-plan bundle.
  • Evidence: 1].abs().sum()) tensor(0., device='cuda:0', dtype=torch.bfloat16) Workaround Make the input tensor contiguous, as in #81665. Related issues #129118 #81665 Additional considerations The fact that the failure is silent...
  • Short excerpt: 1].abs().sum()) tensor(0., device='cuda:0', dtype=torch.bfloat16) Workaround Make the input tensor contiguous, as in #81665. Related issues #129118 #81665 Additional considerations The fact that the failure is silent makes this hard to identify, track and debug. The CPU implem...
  • Review threads: thread:issue:129118:issue:141831, thread:issue:144824:issue:188756:issue:84520, thread:issue:117736:issue:155218:issue:95024, +1
  • Package receipt: ed335031f944

Bundle BUNDLE-1185: module: __torch_dispatch__ stale/low urgency batch 2

Stale issue resolution sweepstale/low urgency41 issues0 PRsmedium
Issues #62888#64548#64874#65339#65423#65619#66026#68622#69530#69991#72179#72911+29
PRs none
  • Diagnosis: This is a module: __torch_dispatch__ stale issue batch, not a single code diagnosis.
  • Next action: Apply one sweep policy: close with rationale when no live reproducer exists, mark canonical duplicates, or promote still-reproducible failures into a local patch-plan bundle.
  • Evidence: issue #81240: stale/low urgency; labels: triaged, module: __torch_dispatch__, module: functionalization; title: Functionalization and fake tensors failure in torture test | issue #74991: stale/low urgency; labels: tri...
  • Short excerpt: issue #81240: stale/low urgency; labels: triaged, module: __torch_dispatch__, module: functionalization; title: Functionalization and fake tensors failure in torture test | issue #74991: stale/low urgency; labels: triaged, better-engineering, module: __torch_dispatch__, tensor...
  • Review threads: thread:issue:69991:issue:72911:issue:84565, thread:issue:77223:issue:77332, thread:issue:124788:issue:84864
  • Package receipt: 90225f4c4ae4

Bundle BUNDLE-1186: module: __torch_function__ stale/low urgency batch 2

Stale issue resolution sweepstale/low urgency34 issues0 PRsmedium
Issues #44316#45565#53678#53937#55095#55104#55440#61544#63911#64187#64690#65215+22
PRs none
  • Diagnosis: This is a module: __torch_function__ stale issue batch, not a single code diagnosis.
  • Next action: Apply one sweep policy: close with rationale when no live reproducer exists, mark canonical duplicates, or promote still-reproducible failures into a local patch-plan bundle.
  • Evidence: issue #76734: stale/low urgency; labels: triaged, module: __torch_function__; title: __torch_function__ callers should always pass kwargs | issue #76294: stale/low urgency; labels: triaged, module: vmap, module: __tor...
  • Short excerpt: issue #76734: stale/low urgency; labels: triaged, module: __torch_function__; title: __torch_function__ callers should always pass kwargs | issue #76294: stale/low urgency; labels: triaged, module: vmap, module: __torch_function__; title: torch._remove_batch_dim is interceptab...
  • Review threads: thread:issue:44316:issue:45565:issue:53678:issue:53937
  • Package receipt: ef97ef3f503d

Bundle BUNDLE-1187: Review thread: 2 issues - Issue #47993 - build pytorch from source on ubuntu, building error from fbgemm::SparseAdaGradSignature

Stale issue resolution sweepstale/low urgency23 issues0 PRsmedium
Issues #2575#17658#28754#41797#47993#51039#51781#55267#60341#61260#62650#64645+11
PRs none
  • Diagnosis: This is a module: abi stale issue batch, not a single code diagnosis.
  • Next action: Apply one sweep policy: close with rationale when no live reproducer exists, mark canonical duplicates, or promote still-reproducible failures into a local patch-plan bundle.
  • Evidence: port torch; print(torch._C._GLIBCXX_USE_CXX11_ABI)"` with the conda version also yields `False` I tried compiling from source but am facing #47993. Any feedback or tips would be appreciated. | ssages sent to stderr or...
  • Short excerpt: port torch; print(torch._C._GLIBCXX_USE_CXX11_ABI)"` with the conda version also yields `False` I tried compiling from source but am facing #47993. Any feedback or tips would be appreciated. | ssages sent to stderr or stdout. Yet, I'm speculating if this crash might be hiding...
  • Review threads: thread:issue:47993:issue:51039, thread:issue:109923:issue:2575, thread:issue:55267:issue:60341:issue:61260:issue:68905
  • Package receipt: 9b40fb8109d2

Bundle BUNDLE-1188: Review thread: 3 issues - Issue #143112 - FFT half precision only let CUDA pass the check

Stale issue resolution sweepstale/low urgency9 issues0 PRsmedium
Issues #143112#150663#165038#166054#166696#167210#167447#169277#169371
PRs none
  • Diagnosis: This is a module: accelerator stale issue batch, not a single code diagnosis.
  • Next action: Apply one sweep policy: close with rationale when no live reproducer exists, mark canonical duplicates, or promote still-reproducible failures into a local patch-plan bundle.
  • Evidence: torch.accelerator.get_device_capability for users to get some device capabilities (currently focus on data types) to solve some issues like #165038 and #143112. The supported data types in PyTorch are: float32, float6...
  • Short excerpt: torch.accelerator.get_device_capability for users to get some device capabilities (currently focus on data types) to solve some issues like #165038 and #143112. The supported data types in PyTorch are: float32, float64, float16, bfloat16, float8_e5m2, float8_e4m3fn, float8_e5m...
  • Review threads: thread:issue:143112:issue:165038:issue:169277
  • Package receipt: 459f88686ca5
269 more bundle receipt(s)

Remaining bundles stay in the downloadable public-safe CSV artifacts to keep this action tab fast to scan.

prepare patch-plan bundle

Volunteer bundles
Why

The local evidence is concrete enough to plan a patch or design branch without writing to GitHub.

Next action

Draft locally, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.

Impact

398 bundles | 4421 issues | 0 PRs
Confidence: medium

Evidence

patch-plan readyissue #185581: has linked PRlabels: triagedtitle: `torch.compile` fails at FakeTensor validation f...398 package receipts

View bundles

Bundle BUNDLE-886: general has linked PR

New code/decision packagehas linked PR34 issues0 PRsmedium
Issues #7887#42652#45808#48972#57724#97635#108784#110944#116882#117032#117962#118263+22
PRs none
  • Diagnosis: This general bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against src/, test/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: issue #185581: has linked PR; labels: triaged; title: `torch.compile` fails at FakeTensor validation for ConvTranspose2d | issue #164478: has linked PR; labels: triaged, actionable; title: Out of date references in CO...
  • Short excerpt: issue #185581: has linked PR; labels: triaged; title: `torch.compile` fails at FakeTensor validation for ConvTranspose2d | issue #164478: has linked PR; labels: triaged, actionable; title: Out of date references in CONTRIBUTING.md and Wiki | issue #130641: has linked PR; label...
  • Package receipt: dbad2d7ed291

Bundle BUNDLE-887: feature needs design decision

New code/decision packageneeds design decision36 issues0 PRsmedium
Issues #32584#32662#37067#38155#39780#42581#44024#46153#49338#49870#51275#53464+24
PRs none
  • Diagnosis: This general bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against src/, test/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: issue #116354: needs design decision; labels: feature, triaged, needs research, needs design; title: Abstract torch.device for GPU/NE/TPU computations in the cloud-based agent | issue #113730: needs design decision; l...
  • Short excerpt: issue #116354: needs design decision; labels: feature, triaged, needs research, needs design; title: Abstract torch.device for GPU/NE/TPU computations in the cloud-based agent | issue #113730: needs design decision; labels: feature, triaged, needs design; title: Can the offici...
  • Package receipt: fd9cd814d7fe

Bundle BUNDLE-888: module: 64-bit has linked PR

New code/decision packagehas linked PR4 issues0 PRsmedium
Issues #119147#131185#167086#170798
PRs none
  • Diagnosis: This module: 64-bit bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: issue #167086: has linked PR; labels: high priority, module: cuda, triaged, module: 64-bit; title: Unchecked uint32 overflow in CUDA scan (cumsum/cumprod/cummin/etc) | issue #170798: has linked PR; labels: module: nn,...
  • Short excerpt: issue #167086: has linked PR; labels: high priority, module: cuda, triaged, module: 64-bit; title: Unchecked uint32 overflow in CUDA scan (cumsum/cumprod/cummin/etc) | issue #170798: has linked PR; labels: module: nn, triaged, module: 64-bit, actionable, module: correctness (s...
  • Package receipt: b287dbb3e360

Bundle BUNDLE-889: module: __torch_dispatch__ needs design decision

New code/decision packageneeds design decision10 issues0 PRsmedium
Issues #62982#73513#73537#79205#96319#97500#113089#117115#132651#140909
PRs none
  • Diagnosis: This module: __torch_dispatch__ bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: issue #132651: needs design decision; labels: triaged, needs design, module: __torch_dispatch__, tensor subclass, oncall: pt2, module: aotdispatch, module: compile-time, module: pt2-dispatcher; title: fix unnecessary...
  • Short excerpt: issue #132651: needs design decision; labels: triaged, needs design, module: __torch_dispatch__, tensor subclass, oncall: pt2, module: aotdispatch, module: compile-time, module: pt2-dispatcher; title: fix unnecessary re-trace in AOTDispatcher in subclass code path | issue #792...
  • Review threads: thread:issue:96319:issue:97500, thread:issue:73513:issue:73537
  • Package receipt: 95b25f401b1e

Bundle BUNDLE-890: module: __torch_function__ has linked PR

New code/decision packagehas linked PR14 issues0 PRsmedium
Issues #47117#54881#58212#63485#96579#103752#115741#147995#167223#182318#186203#186259+2
PRs none
  • Diagnosis: This module: __torch_function__ bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: issue #167223: has linked PR; labels: module: docs, module: autograd, triaged, actionable, module: __torch_function__; title: [feature request] Re-enable `__torch_function__` dispatch when invoking autograd hooks for...
  • Short excerpt: issue #167223: has linked PR; labels: module: docs, module: autograd, triaged, actionable, module: __torch_function__; title: [feature request] Re-enable `__torch_function__` dispatch when invoking autograd hooks for Tensor subclasses | issue #182318: has linked PR; labels: tr...
  • Review threads: thread:issue:47117:issue:58212:issue:63485
  • Package receipt: 505c70e927db

Bundle BUNDLE-891: module: abi needs maintainer decision

New code/decision packageneeds maintainer decision4 issues0 PRsmedium
Issues #103359#121837#124197#174507
PRs none
  • Diagnosis: This module: abi bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: issue #174507: needs maintainer decision; labels: feature, module: abi, triaged, bot-triaged; title: PyTorch Stable ABI support | issue #121837: needs maintainer decision; labels: module: binaries, module: cpp, module...
  • Short excerpt: issue #174507: needs maintainer decision; labels: feature, module: abi, triaged, bot-triaged; title: PyTorch Stable ABI support | issue #121837: needs maintainer decision; labels: module: binaries, module: cpp, module: abi, triaged; title: Revive libtorch-shared-without-deps f...
  • Package receipt: e1c23fc4dc50

Bundle BUNDLE-892: Review thread: 2 issues - Issue #175211 - CUDA/ROCm/Accelerator testing should replace get_device_capability() with feature queries

New code/decision packagehas linked PR12 issues0 PRsmedium
Issues #143887#156805#158827#164302#175211#177010#177771#177935#177969#185975#186260#189014
PRs none
  • Diagnosis: This module: accelerator bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: I agreed, and a similar idea from #175211 | issue #185975: has linked PR; labels: feature, triaged, module: PrivateUse1, module: accelerator, bot-triaged; title: torch.accelerator.default_stream API is missing | our c...
  • Short excerpt: I agreed, and a similar idea from #175211 | issue #185975: has linked PR; labels: feature, triaged, module: PrivateUse1, module: accelerator, bot-triaged; title: torch.accelerator.default_stream API is missing | our categories in a backend-agnostic way. And generalize the adva...
  • Review threads: thread:issue:175211:issue:177969, thread:issue:158827:issue:164302:issue:177010
  • Package receipt: 580e6eecb1da

Bundle BUNDLE-893: Review thread: 3 issues - Issue #180343 - Accelerator-independent API to determine if a specific backend is available?

New code/decision packageneeds maintainer decision6 issues0 PRsmedium
Issues #166563#167283#167881#180343#180345#180712
PRs none
  • Diagnosis: This module: accelerator bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: 🚀 The feature, motivation and pitch This probably lands somewhere between #180343 and #180345, not sure if it is justified in having its own issue, but in TorchTPU we have set up autoloading of our tpu module, but we...
  • Short excerpt: 🚀 The feature, motivation and pitch This probably lands somewhere between #180343 and #180345, not sure if it is justified in having its own issue, but in TorchTPU we have set up autoloading of our tpu module, but we only | 🚀 The feature, motivation and pitch This probably lan...
  • Review threads: thread:issue:180343:issue:180345:issue:180712, thread:issue:166563:issue:167283:issue:167881
  • Package receipt: 731017ff4f38

Bundle BUNDLE-894: Review thread: 7 issues - Issue #113045 - [Feature][DTensor] Manage additional `_padded_local_tensor` attribute

New code/decision packageneeds maintainer decision17 issues0 PRsmedium
Issues #113045#113737#114299#120003#120945#121594#129867#136433#147662#155547#158657#163923+5
PRs none
  • Diagnosis: This module: activation checkpointing bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: willing to work on this one here is one related issue #158657 | more. This is the natural continuation of the per-parameter FSDP direction (RFC #114299) and is adjacent to the activation-offload work in #174960, on th...
  • Short excerpt: willing to work on this one here is one related issue #158657 | more. This is the natural continuation of the per-parameter FSDP direction (RFC #114299) and is adjacent to the activation-offload work in #174960, on the parameter axis rather than the activation axis. Proposed A...
  • Review threads: thread:issue:113045:issue:114299:issue:120003:issue:147662
  • Package receipt: 75aeea97036e

Bundle BUNDLE-895: #180855 `"index_copy_cpu"` not implemented for `'Float8_e5m2'`

New code/decision packagehas linked PR4 issues0 PRsmedium
Issues #110946#116555#180855#189046
PRs none
  • Diagnosis: This module: advanced indexing bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: issue #180855: has linked PR; labels: module: cpu, triaged, enhancement, module: advanced indexing, module: floatx (formerly float8), bot-triaged; title: `"index_copy_cpu"` not implemented for `'Float8_e5m2'` | issue...
  • Short excerpt: issue #180855: has linked PR; labels: module: cpu, triaged, enhancement, module: advanced indexing, module: floatx (formerly float8), bot-triaged; title: `"index_copy_cpu"` not implemented for `'Float8_e5m2'` | issue #189046: needs maintainer decision; labels: feature, module:...
  • Package receipt: 5158149500a8

Bundle BUNDLE-896: Review thread: 15 issues - Issue #104247 - [proposal] "Name" string attribute for modules, parameters, buffers, tensors for more pleasant debugging (including distributed FSDP2 logging)

New code/decision packageneeds design decision26 issues0 PRsmedium
Issues #1529#16668#25267#27479#30049#31252#31353#35633#45273#47128#58228#58392+14
PRs none
  • Diagnosis: This module: amp (automated mixed precision) bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: This is a very close duplicate of #31252, and there is a discussion going on already there. TL;DR it's complicated. | Related #31252 | @eellison My impression is that this is unrelated to #25267 given that I am not us...
  • Short excerpt: This is a very close duplicate of #31252, and there is a discussion going on already there. TL;DR it's complicated. | Related #31252 | @eellison My impression is that this is unrelated to #25267 given that I am not using MKL and I only encounter the CPU memory leak when the mo...
  • Review threads: thread:issue:104247:issue:1529:issue:16668:issue:167656
  • Package receipt: 9069be9c0a49

Bundle BUNDLE-897: Review thread: 15 issues - Issue #102078 - AddressSanitizer: heap-buffer-overflow in test_comprehensive_nn_functional_embedding_bag_cpu_bfloat16

New code/decision packageneeds reproduction16 issues0 PRsmedium
Issues #88#13993#14436#19668#23240#48491#66247#70245#89817#95944#99561#102078+4
PRs none
  • Diagnosis: This module: android bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: I have also seen the same issue in #66247 and had to resort to building my own PyTorch container with `USE_MKL=OFF` in order to resolve this issue. My stack trace was also similar t | 000004ea137 in PyCFunction_Call (...
  • Short excerpt: I have also seen the same issue in #66247 and had to resort to building my own PyTorch container with `USE_MKL=OFF` in order to resolve this issue. My stack trace was also similar t | 000004ea137 in PyCFunction_Call () at ../Objects/methodobject.c:98 #87 0x00000000005c20e7 in...
  • Review threads: thread:issue:102078:issue:123421:issue:126963:issue:13993
  • Package receipt: 11f6973c70da
386 more bundle receipt(s)

Remaining bundles stay in the downloadable public-safe CSV artifacts to keep this action tab fast to scan.

prepare volunteer-ready bundle

Volunteer bundles
Why

The grouped issues share low-risk scope, labels, or documentation/test expectations suitable for contributors.

Next action

Publish or keep a local contributor-ready bundle with scope, expected files, tests, and linked issue context.

Impact

15 bundles | 22 issues | 0 PRs
Confidence: medium

Evidence

patch-plan readyissue #117806: good volunteer slicelabels: good first issue, triaged, enhancement, module:...title: Supporting custom attributes with `__torch_funct...15 package receipts

View bundles

Bundle BUNDLE-1162: #117806 Supporting custom attributes with `__torch_function__` tensor subclasses

New volunteer-ready PRgood volunteer slice1 issue0 PRsmedium
Issues #117806
PRs none
  • Diagnosis: This module: __torch_function__ bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: issue #117806: good volunteer slice; labels: good first issue, triaged, enhancement, module: __torch_function__, module: dynamo; title: Supporting custom attributes with `__torch_function__` tensor subclasses
  • Short excerpt: issue #117806: good volunteer slice; labels: good first issue, triaged, enhancement, module: __torch_function__, module: dynamo; title: Supporting custom attributes with `__torch_function__` tensor subclasses
  • Package receipt: d846dd808be3

Bundle BUNDLE-1163: module: bootcamp good volunteer slice

New volunteer-ready PRgood volunteer slice2 issues0 PRsmedium
Issues #58122#79452
PRs none
  • Diagnosis: This module: bootcamp bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: issue #58122: good volunteer slice; labels: module: bootcamp, good first issue, triaged, OSS contribution wanted; title: Generate special clear error messages for known common misuses in TorchScript | issue #79452: go...
  • Short excerpt: issue #58122: good volunteer slice; labels: module: bootcamp, good first issue, triaged, OSS contribution wanted; title: Generate special clear error messages for known common misuses in TorchScript | issue #79452: good volunteer slice; labels: oncall: distributed, module: boo...
  • Package receipt: da0fb0d4012d

Bundle BUNDLE-1164: #99278 Build error on libstc++ header stl_alogbase.h on riscv

New volunteer-ready PRgood volunteer slice1 issue0 PRsmedium
Issues #99278
PRs none
  • Diagnosis: This module: build bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: issue #99278: good volunteer slice; labels: module: build, good first issue, triaged; title: Build error on libstc++ header stl_alogbase.h on riscv
  • Short excerpt: issue #99278: good volunteer slice; labels: module: build, good first issue, triaged; title: Build error on libstc++ header stl_alogbase.h on riscv
  • Package receipt: e59d59de9091

Bundle BUNDLE-1165: module: c10d good volunteer slice

New volunteer-ready PRgood volunteer slice2 issues0 PRsmedium
Issues #84593#97469
PRs none
  • Diagnosis: This module: c10d bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: issue #97469: good volunteer slice; labels: good first issue, triaged, module: c10d; title: Improve collectives fingerprinting | issue #84593: good volunteer slice; labels: good first issue, triaged, module: c10d; tit...
  • Short excerpt: issue #97469: good volunteer slice; labels: good first issue, triaged, module: c10d; title: Improve collectives fingerprinting | issue #84593: good volunteer slice; labels: good first issue, triaged, module: c10d; title: Uneven and/or Dynamically sized collectives
  • Package receipt: 7b39e91240ea

Bundle BUNDLE-1166: Review thread: 2 issues - Issue #64067 - Execute C++ based GTest unit tests from a python wrapper

New volunteer-ready PRgood volunteer slice2 issues0 PRsmedium
Issues #64067#66824
PRs none
  • Diagnosis: This module: ci bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: #65514 introduced a Python framework for running C++ tests(the benefits and reasoning are described in #64067). While that PR added a framework for running tests, we still have many manual calls to C++ test binaries i...
  • Short excerpt: #65514 introduced a Python framework for running C++ tests(the benefits and reasoning are described in #64067). While that PR added a framework for running tests, we still have many manual calls to C++ test binaries in the scripts used in CI. This i
  • Review threads: thread:issue:64067:issue:66824
  • Package receipt: dd1bc8ca2b8f

Bundle BUNDLE-1167: #153669 Make tlparse able to show a summary of distinct graph breaks

New volunteer-ready PRgood volunteer slice1 issue0 PRsmedium
Issues #153669
PRs none
  • Diagnosis: This module: compile ux bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: issue #153669: good volunteer slice; labels: good first issue, triaged, oncall: pt2, module: dynamo, dynamo-triage-jan2025, module: compile ux; title: Make tlparse able to show a summary of distinct graph breaks
  • Short excerpt: issue #153669: good volunteer slice; labels: good first issue, triaged, oncall: pt2, module: dynamo, dynamo-triage-jan2025, module: compile ux; title: Make tlparse able to show a summary of distinct graph breaks
  • Package receipt: b998105f7b5b

Bundle BUNDLE-1168: #9886 [doc] functionalities not documented

New volunteer-ready PRgood volunteer slice1 issue0 PRsmedium
Issues #9886
PRs none
  • Diagnosis: This module: docs bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: issue #9886: good volunteer slice; labels: module: docs, good first issue, triaged; title: [doc] functionalities not documented
  • Short excerpt: issue #9886: good volunteer slice; labels: module: docs, good first issue, triaged; title: [doc] functionalities not documented
  • Package receipt: 3b48f90fa778

Bundle BUNDLE-1169: Review thread: 2 issues - Issue #102207 - skipIfTorchInductor Tracking Issue

New volunteer-ready PRgood volunteer slice2 issues0 PRsmedium
Issues #102207#108798
PRs none
  • Diagnosis: This module: dynamo bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: _.from needs to be renamed" 'test_advancedindex_mixed_cpu_devices' - "FIXME" 'test_advancedindex_mixed_devices_error' - "FIXME" #108181 and #108798 'test_typed_storage_deprecation_warning' - "FIXME" 'test_pin_memory'...
  • Short excerpt: _.from needs to be renamed" 'test_advancedindex_mixed_cpu_devices' - "FIXME" 'test_advancedindex_mixed_devices_error' - "FIXME" #108181 and #108798 'test_typed_storage_deprecation_warning' - "FIXME" 'test_pin_memory' - "pin_memory isn't yet supported in TorchInductor" 'test_me...
  • Review threads: thread:issue:102207:issue:108798
  • Package receipt: e8ae53853c07

Bundle BUNDLE-1170: module: internals good volunteer slice

New volunteer-ready PRgood volunteer slice3 issues0 PRsmedium
Issues #10756#20704#61654
PRs none
  • Diagnosis: This module: internals bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: issue #61654: good volunteer slice; labels: module: internals, good first issue, triaged; title: Allow creation of pseudo devices for testing purposes | issue #10756: good volunteer slice; labels: module: internals, g...
  • Short excerpt: issue #61654: good volunteer slice; labels: module: internals, good first issue, triaged; title: Allow creation of pseudo devices for testing purposes | issue #10756: good volunteer slice; labels: module: internals, good first issue, triaged, module: partial aliasing; title: U...
  • Package receipt: b99cd5d3da96

Bundle BUNDLE-1171: Review thread: 2 issues - Issue #40316 - Refactor the adaptive avg pool code

New volunteer-ready PRgood volunteer slice2 issues0 PRsmedium
Issues #40316#42779
PRs none
  • Diagnosis: This module: memory format bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: ng5 @Xia-Weiwen @leslie-fang-intel @dzhulgakov @VitalyFedyunin @ppwwyyxx @ngimel @z-a-f, @vkuzo. 🐛 Bug As part of the refactoring effort in #40316, I ran some benchmarks on adaptive_avg_pool(2/3)d to see what is the d...
  • Short excerpt: ng5 @Xia-Weiwen @leslie-fang-intel @dzhulgakov @VitalyFedyunin @ppwwyyxx @ngimel @z-a-f, @vkuzo. 🐛 Bug As part of the refactoring effort in #40316, I ran some benchmarks on adaptive_avg_pool(2/3)d to see what is the difference in execution time for MemoryFormat::Contiguous vs...
  • Review threads: thread:issue:40316:issue:42779
  • Package receipt: c8c38eba8b96

Bundle BUNDLE-1172: #66813 Use `__slots__` for the `nn.Module` class

New volunteer-ready PRgood volunteer slice1 issue0 PRsmedium
Issues #66813
PRs none
  • Diagnosis: This module: nn bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: issue #66813: good volunteer slice; labels: module: performance, module: nn, good first issue, triaged, better-engineering, actionable; title: Use `__slots__` for the `nn.Module` class
  • Short excerpt: issue #66813: good volunteer slice; labels: module: performance, module: nn, good first issue, triaged, better-engineering, actionable; title: Use `__slots__` for the `nn.Module` class
  • Package receipt: 871dc27de148

Bundle BUNDLE-1173: #93684 Can we rewrite numpy operators to pytorch operators?

New volunteer-ready PRgood volunteer slice1 issue0 PRsmedium
Issues #93684
PRs none
  • Diagnosis: This module: numpy bundle is concrete enough for a local patch-plan bundle.
  • Next action: Draft locally against torch/, test/, .ci/, add focused regression coverage, and only open a GitHub PR if maintainers approve the scope.
  • Evidence: issue #93684: good volunteer slice; labels: good first issue, triaged, module: numpy; title: Can we rewrite numpy operators to pytorch operators?
  • Short excerpt: issue #93684: good volunteer slice; labels: good first issue, triaged, module: numpy; title: Can we rewrite numpy operators to pytorch operators?
  • Package receipt: 06df091b836c
3 more bundle receipt(s)

Remaining bundles stay in the downloadable public-safe CSV artifacts to keep this action tab fast to scan.