Skip to content

Issues: pytorch/pytorch

[v2.4.1] Release Tracker
#132400 opened Aug 1, 2024 by atalman
Open 7
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

13,687 Open 30,400 Closed
13,687 Open 30,400 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Number of kernels generated increased after changes to statically_known_multiple_of oncall: pt2
#133361 opened Aug 13, 2024 by henrylhtsang
[Dynamo] Map/zip IteratorVariable updates causing graph breaks dynamo-variable-tracker module: dynamo needs reproduction Someone else needs to try reproducing the issue given the instructions. No action needed from user oncall: pt2 triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#133358 opened Aug 13, 2024 by williamwen42
@williamwen42
arange with integer end only produces complicated length expression module: dynamic shapes oncall: pt2
#133338 opened Aug 13, 2024 by ezyang
_suggest_torch_checks: TypeError: sequence item 0: expected str instance, Symbol found oncall: export
#133336 opened Aug 13, 2024 by ezyang
1
@avikchaudhuri
1
torch._scaled_mm row-wise hits CUDA invalid memory access when M % 256 != 0
#133334 opened Aug 13, 2024 by lw
1
[CI] raw logs with color are unreadable actionable module: ci Related to continuous integration module: regression It used to work, and now it doesn't needs reproduction Someone else needs to try reproducing the issue given the instructions. No action needed from user triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#133330 opened Aug 13, 2024 by zou3519
2
Flexattention: ValueError: Shape element 1 must be a power of 2
#133321 opened Aug 13, 2024 by foreverpiano
3
Update python custom ops tutorial per user feedback module: pt2-dispatcher PT2 dispatcher-related issues (e.g., aotdispatch, functionalization, faketensor, custom-op, oncall: pt2
#133318 opened Aug 13, 2024 by zou3519
@zou3519
RuntimeError: INTERNAL ASSERT FAILED CUDACachingAllocator.cpp
#133316 opened Aug 13, 2024 by manu-chauhan
DISABLED test_b2b_gemm_right_assoc_good_shape (__main__.B2BGEMMTest) module: rocm AMD GPU support for Pytorch skipped Denotes a (flaky) test currently skipped in CI.
#133311 opened Aug 13, 2024 by jataylo
1
Contributor pipeline ignores GitHub-co-authorship module: ci Related to continuous integration triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#133310 opened Aug 13, 2024 by 1ucian0
@malfet
1
Torch Profiler distributed info is not JSON serializable
#133308 opened Aug 13, 2024 by anaprietonem
The performance regression of conv_transpose1d on CPU with variable-shaped inputs
#133304 opened Aug 13, 2024 by Sytronik
@yanbing-j
Graph break when there is backward hook registered for sub module
#133303 opened Aug 13, 2024 by jerrychenhf
Implementing caching under torch.compile cause error dynamo-variable-tracker module: dynamo oncall: pt2 triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#133302 opened Aug 13, 2024 by Chillee
@yanboliang
2
why PyTorch relies on a specific commit ID of libgomp.so.1?(I have solved the issue, but I have some questions.) module: binaries Anything related to official binaries that we release to users oncall: pt2
#133294 opened Aug 13, 2024 by sujuyu
CMake Error: Failed to find nvToolsExt
#133293 opened Aug 13, 2024 by efsotr
Give PrivateUse2 and PrivateUse3 the same treatment as PrivateUse1
#133288 opened Aug 13, 2024 by tehbone
DISABLED test_autograd_cpp_node_saved_int (__main__.TestCompiledAutograd) module: flaky-tests Problem is a flaky test in CI module: inductor module: macos Mac OS related issues oncall: pt2 skipped Denotes a (flaky) test currently skipped in CI. triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
#133283 opened Aug 13, 2024 by pytorch-bot bot
4
Make CUDAPluggableAllocator support getDeviceStats, resetPeakStats
#133281 opened Aug 12, 2024 by VibhuJawa
Pluggable Allocator Does not raise torch.cuda.OutOfMemoryError
#133280 opened Aug 12, 2024 by VibhuJawa
Could NOT find CUDA (missing: CUDA_CUDART_LIBRARY) (found version "12.5")
#133275 opened Aug 12, 2024 by Geremia
1
Fakeification fails through combination of subclass, .grad, and ._base having different ndims module: aotdispatch umbrella label for AOTAutograd issues module: fakeTensor module: pt2-dispatcher PT2 dispatcher-related issues (e.g., aotdispatch, functionalization, faketensor, custom-op, oncall: pt2 tensor subclass Related to tensor subclasses
#133274 opened Aug 12, 2024 by bdhirsh
Adafactor compile support tracker module: optimizer Related to torch.optim oncall: pt2
#133268 opened Aug 12, 2024 by janeyx99
3 tasks
torch.ldexp incorrectly returns infinity for float64 inputs larger than float32 max
#133267 opened Aug 12, 2024 by roman-openai
ProTip! Find all open issues with in progress development work with linked:pr.