-
Notifications
You must be signed in to change notification settings - Fork 29k
All issues
Issue creation is restricted in this repository
- #193034 · atalman opened
on Aug 11, 2026 42
Issues
is:issue state:open
is:issue state:open
Search results
[torch.compile][Inductor] Silent wrong results when an automatically dynamic Python float used by clamp reuses a stale value
bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage botStatus: Open.#194976 In pytorch/pytorch;[inductor] SIMDKernel._split_iteration_ranges incorrectly asserts guarding_hint_or_throw(s) == 1 on unbacked u0
bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage bottriagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleStatus: Open.#194972 In pytorch/pytorch;CAN_COMPILE_METAL_40 probe reports success on toolchains that cannot compile the Metal 4 kernels it gates, breaking source builds
bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage botmodule: buildBuild system issuesBuild system issuesmodule: mpsRelated to Apple Metal Performance Shaders frameworkRelated to Apple Metal Performance Shaders frameworktriagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleStatus: Open.#194946 In pytorch/pytorch;scikit-build does not play nice with a large agent_space.
bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage botmodule: buildBuild system issuesBuild system issuesmodule: performanceIssues related to performance, either of kernel code or framework glueIssues related to performance, either of kernel code or framework gluetriagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleStatus: Open.#194945 In pytorch/pytorch;MultiProcContinuousTest hangs forever on stalled ranks
bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage botoncall: distributedAdd this issue/PR to distributed oncall triage queueAdd this issue/PR to distributed oncall triage queueoncall: distributed infraAdd this issue/PR to distributed infra oncall triage queueAdd this issue/PR to distributed infra oncall triage queuetriagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleStatus: Open.#194943 In pytorch/pytorch;Greenlight User Guide
bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage botStatus: Open.#194928 In pytorch/pytorch;Gamma distribution broken on Intel GPUs
bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage botmodule: correctness (silent)issue that returns an incorrect result silentlyissue that returns an incorrect result silentlymodule: crashProblem manifests as a hard crash, as opposed to a RuntimeErrorProblem manifests as a hard crash, as opposed to a RuntimeErrormodule: distributionsRelated to torch.distributionsRelated to torch.distributionsmodule: xpuIntel XPU related issuesIntel XPU related issuesStatus: Open.#194926 In pytorch/pytorch;F.pad(mode="constant") with temporal-axis padding silently corrupts ~80% of elements for large 5D tensors on MPS
bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage botmodule: correctness (silent)issue that returns an incorrect result silentlyissue that returns an incorrect result silentlymodule: mpsRelated to Apple Metal Performance Shaders frameworkRelated to Apple Metal Performance Shaders frameworkStatus: Open.#194922 In pytorch/pytorch;get_attrnode warning when using nestedtorch.func.linearizebot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage botmodule: functorchPertaining to torch.func or pytorch/functorchPertaining to torch.func or pytorch/functorchtriagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleStatus: Open.#194903 In pytorch/pytorch;DISABLED test_layout_TT_dynamic_k (__main__.LayoutCoverageWildcardTest)
module: flaky-testsProblem is a flaky test in CIProblem is a flaky test in CIskippedDenotes a (flaky) test currently skipped in CI.Denotes a (flaky) test currently skipped in CI.triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleStatus: Open.#194900 In pytorch/pytorch;DISABLED test_layout_TN_dynamic_k (__main__.LayoutCoverageWildcardTest)
module: flaky-testsProblem is a flaky test in CIProblem is a flaky test in CIskippedDenotes a (flaky) test currently skipped in CI.Denotes a (flaky) test currently skipped in CI.triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleStatus: Open.#194899 In pytorch/pytorch;[TorchInductor] normalization_pass misses torch.concatenate
bot-triagedThis is a label only to be used by the auto triage botThis is a label only to be used by the auto triage bottriagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleStatus: Open.#194898 In pytorch/pytorch;
