id int64 393k 2.82B | repo stringclasses 68
values | title stringlengths 1 936 | body stringlengths 0 256k ⌀ | labels stringlengths 2 508 | priority stringclasses 3
values | severity stringclasses 3
values |
|---|---|---|---|---|---|---|
1,222,848,892 | pytorch | [feature request] no-param sort to exploit parallelization | ### 🚀 The feature, motivation and pitch
'torch.sort' can not execute parallel when there is only one dimention of tensor. I hope you could make it faster when we try to sort 1 dimention tensors.
### Alternatives
Maybe you could implement merge sort or other parallel sorting algorithms
### Additional cont... | module: performance,module: cpu,triaged,module: multithreading,module: sorting and selection | low | Minor |
1,222,868,267 | terminal | [Settings - Startup]: Keyboard focus is going on 'Startup' heading which is a non-interactive control. | ### Windows Terminal version
1.13.10984.0
### Windows build number
22598.200
### Other Software
**Test Environment:**
**OS:** Windows 11 Version 21H2 (OS Build 22598.200)
**Screen Reader:** Narrator.
### Steps to reproduce
**Repro Steps:**
Open Windows Terminal.
Open Settings page using "Ctrl +... | Issue-Bug,Area-Accessibility,Product-Terminal,Priority-3,HCL-E+D,A11yWCAG,HCL-WindowsTerminal,A11yCO,A11ySev3 | low | Major |
1,222,897,927 | pytorch | Clarify dependency on NumPy (related to maskedtensor?) | ### 🐛 Describe the bug
I installed torch cpu version on a new python3.8 ubuntu 20.04 system via pip. From what I understood, it did not install NumPy as dependency.
When I import torch, I get:
```
/usr/local/lib/python3.8/dist-packages/torch/_masked/__init__.py:223: UserWarning: Failed to initialize NumPy: numpy... | triaged | low | Critical |
1,222,900,804 | vscode | Keep the current reference search results and open new reference in new view | <!-- ⚠️⚠️ Do Not Delete This! feature_request_template ⚠️⚠️ -->
<!-- Please read our Rules of Conduct: https://opensource.microsoft.com/codeofconduct/ -->
<!-- Please search existing issues to avoid creating duplicates. -->
<!-- Describe the feature you'd like. -->
Visual studio has this feature described in http... | feature-request,references-viewlet | medium | Major |
1,222,997,150 | pytorch | Remove all docstrings when python is running in optimization mode | ## Issue description
Follow up of #76034.
There are docstrings still present in *pyc files when python is run in optimization mode (`-OO`).
## Code example
```python
$ /usr/local/bin/python3.9 -OO
Python 3.9.2 (default, Mar 26 2021, 21:58:27)
[GCC 7.5.0] on linux
Type "help", "copyright", "credits" or "... | module: performance,module: docs,triaged,better-engineering,module: python frontend | low | Major |
1,223,002,432 | tauri | High memory usage when invoking commands [bug] | ### Describe the bug
High memory usage when sending lots of commands for rust from js.
### Reproduction
Invoke a command in an interval:
```js
setInterval(() => {
invoke<number[][]>('get_data', args).then(drawOnHtmlCanvasElement)
}, 50)
```
Return large amount of data from rust to js:
```rs
#[tauri... | type: bug | high | Critical |
1,223,019,265 | TypeScript | refactor import statements to import more specific paths | <!-- ⚠️⚠️ Do Not Delete This! feature_request_template ⚠️⚠️ -->
<!-- Please read our Rules of Conduct: https://opensource.microsoft.com/codeofconduct/ -->
<!-- Please search existing issues to avoid creating duplicates. -->
<!-- Describe the feature you'd like. -->
It would be great to have the ability to refac... | Suggestion,Awaiting More Feedback,Domain: Refactorings | low | Major |
1,223,065,852 | ant-design | The size of the button group, which affects the size of the buttons in the Modal. | ### Reproduction link
[https://bzgjxw.csb.app/](https://bzgjxw.csb.app/)
### Steps to reproduce
<img width="883" alt="image" src="https://user-images.githubusercontent.com/69514654/166267474-c0e85658-aec4-41d3-bb6f-0aa80bf1d3bc.png">
```tsx
<Dropdown.Button
size="small"
overlay={<Menu items={[{ key: ... | Inactive | low | Minor |
1,223,086,542 | go | x/website: update `go list -m` documentation for workspace mode in ref/mod | The documentation of `go list -m` currently says:
"If no arguments are specified, the *main module* is listed."
This is incorrect, since, when workspace mode is enabled, `go list -m` lists all the active main modules.
Here is an example:
`$ go list -m -json`
```
{
"Path": "xxx",
"Main": true,
"Dir": "/t... | Documentation,NeedsInvestigation,GoCommand,website | low | Minor |
1,223,089,428 | flutter | When setting flutter config --build-dir=<out>, the default build folder is still created. | <!-- Thank you for using Flutter!
If you are looking for support, please check out our documentation
or consider asking a question on Stack Overflow:
* https://flutter.dev/
* https://api.flutter.dev/
* https://stackoverflow.com/questions/tagged/flutter?sort=frequent
If you hav... | tool,customer: google,a: build,has reproducible steps,P2,found in release: 3.3,found in release: 3.4,team-tool,triaged-tool | low | Critical |
1,223,112,039 | transformers | Collection of Tokenizer issues | ### System Info
```shell
Transformers + Tokenizers
```
### Who can help?
This Issue is a summary of multiple problems that we are currently encountering with Tokenizers. To solve them we'll need a more profound discussion of:
- To what extend fast and slow tokenizers should be aligned
- Whether all slow tokenizer... | Discussion,WIP,bug | low | Major |
1,223,143,989 | pytorch | C++ CUDA assign existing memory to forward method. | ### 🐛 Describe the bug
Apologies for posting this here. I did post a question in the forum however no one replied for a couple of weeks. I would like to load a pytorch model (which Is trained in python) in a C++ application. The application has already performed device allocations and is quite memory intensive. So, I... | oncall: jit | low | Critical |
1,223,156,584 | go | cmd/compile: SIGILL when cross-compiling from netbsd/arm64 to linux/mips | ```
# internal/abi
SIGILL: illegal instruction
PC=0x273d40 m=4 sigcode=4
instruction bytes: 0xa8 0x74 0x41 0x79 0x8 0x1 0x7f 0xb2 0xa8 0x74 0x1 0x79 0xd2 0xf9 0xff 0x17
goroutine 9 [running]:
cmd/internal/obj/mips.preprocess(0x4... | OS-NetBSD,NeedsInvestigation,arch-arm64,compiler/runtime | low | Critical |
1,223,162,715 | vscode | Local terminal can't be opened: \bin\bash not found (though VS Code running on Windows) | <!-- Please search existing issues to avoid creating duplicates, and review our troubleshooting tips: https://code.visualstudio.com/docs/remote/troubleshooting -->
<!-- Please attach logs to help us diagnose your issue. Learn more here: https://code.visualstudio.com/docs/remote/troubleshooting#_reporting-issues -->
<... | bug,remote,confirmed,terminal-process | medium | Critical |
1,223,168,858 | go | runtime: make underlying implementation of channels more efficient (Reference -> https://github.com/alphadose/ZenQ) | Current implementation of channels could use some improvements in terms of `ns/op`, `B/op` and `allocs/op`
I have made a POC thread-safe queue which has better performance in all the above 3 metrics [https://github.com/alphadose/ZenQ](https://github.com/alphadose/ZenQ)
Here are the benchmarks vs native channels [... | Performance,NeedsInvestigation | high | Major |
1,223,172,122 | angular | Enable the `no-duplicate-enum-values` TSLint rule for the FW repo | ### Which @angular/* package(s) are relevant/related to the feature request?
core
### Description
As a followup of https://github.com/angular/angular/pull/45825#discussion_r862797290, we should consider enabling `no-duplicate-enum-values` TSLint rule to catch accidental duplicate enum values. | feature,area: build & ci,feature: in backlog | low | Minor |
1,223,189,531 | rust | Random `lifetime may not live long enough` from other part of the code | I just stumbled across a weird and inconsistent behaviour, that seems unreasonable to me.
The project I'm working on is a big company project (closed source) with quite some files. While fixing a bug and editing one line in one module (`src/dispatcher/compiler/module_range.rs`) the compiler suddenly decided to rejec... | P-medium,T-compiler,A-incr-comp,A-NLL,NLL-complete | medium | Critical |
1,223,207,290 | neovim | build: write minimal logs to stdout, full log as CI artifact | ## Problem
Build logs in CI are mostly noise. But sometimes we need the full details of the log.
https://github.com/neovim/neovim/pull/17906#issuecomment-1109171265
## Solution
1. Modify `test/busted/outputHandlers/nvim.lua` so that:
1. when running in CI, it writes the full logs (as today) to a known ... | enhancement,build,test,ci | low | Critical |
1,223,215,894 | pytorch | [torch::deploy] Remove manager_ from AquireSession | [torch::deploy](https://github.com/pytorch/pytorch/blob/master/docs/source/deploy.rst) is a runtime for getting pytorch models to run on multiple threads in eager mode. There's a great whitepaper about torch.package and torch::deploy [here](https://arxiv.org/pdf/2104.00254.pdf).
Motivation: We want to be able to bre... | oncall: package/deploy,imported | low | Critical |
1,223,217,540 | go | spec: provide some way to refer to a generic function without instantiation | The motivation behind type aliases was the ability to refer under a new name to a thing that exists elsewhere. Prior to generics, functions can be easily re-referred to using assignment:
```go
var Cut = strings.Cut
```
However, doing the same thing with a generic function fails with error message "cannot use ge... | NeedsInvestigation,generics | medium | Critical |
1,223,220,415 | pytorch | [torch::deploy] remove reliance on manager_ for unload | [torch::deploy](https://github.com/pytorch/pytorch/blob/master/docs/source/deploy.rst) is a runtime for getting pytorch models to run on multiple threads in eager mode. There's a great whitepaper about torch.package and torch::deploy [here](https://arxiv.org/pdf/2104.00254.pdf).
Motivation: We want to be able to bre... | oncall: package/deploy,imported | low | Critical |
1,223,227,831 | PowerToys | Child/pop up window snapping plus centering in zone | ### Description of the new feature / enhancement
Add option to center child windows/popups in the zone vs snapping/maximizing. Some windows cannot maximize and are simply snapped to the upper left corner.
### Scenario when this would be used?
Some child/pop up windows cannot be maximized (example: Outlook Notificat... | Idea-Enhancement,Product-FancyZones,Needs-Triage | low | Minor |
1,223,233,851 | pytorch | [torch::deploy] move create_movable to interpreter_manager | [torch::deploy](https://github.com/pytorch/pytorch/blob/master/docs/source/deploy.rst) is a runtime for getting pytorch models to run on multiple threads in eager mode. There's a great whitepaper about torch.package and torch::deploy [here](https://arxiv.org/pdf/2104.00254.pdf).
Motivation: We want to be able to bre... | oncall: package/deploy,imported | low | Critical |
1,223,235,374 | pytorch | [torch::deploy] Remove `manager_` from the constructor and deconstructor of `InterpreterSession` | [torch::deploy](https://github.com/pytorch/pytorch/blob/master/docs/source/deploy.rst) is a runtime for getting pytorch models to run on multiple threads in eager mode. There's a great whitepaper about torch.package and torch::deploy [here](https://arxiv.org/pdf/2104.00254.pdf).
Motivation: We want to be able to bre... | module: deploy,oncall: package/deploy,imported | low | Critical |
1,223,236,010 | flutter | Vectorize `BoxConstraints` | The following is the profile for `layout_bench.dart` in Flutter. A not insignificant amount of time is spent in `Double.clamp`. This should be vectorized. Furthermore in `customer:money` layout time takes up ~~450ms~~ 200ms of startup time on low end android devices, so there is a lot of opportunity here.
<img wid... | engine,dependency: dart,c: proposal,P3,team-engine,triaged-engine | low | Major |
1,223,245,838 | go | x/net/http2: apparent deadlock in TestTransportResponseHeaderTimeout_Body on windows-386-2008 | ```
panic: test timed out after 10m0s
…
goroutine 10660 [semacquire, 9 minutes]:
sync.runtime_Semacquire(0x13397808)
C:/workdir/go/src/runtime/sema.go:56 +0x36
sync.(*WaitGroup).Wait(0x13397800)
C:/workdir/go/src/sync/waitgroup.go:136 +0x87
golang.org/x/net/http2.(*clientTester).run(0x130e9110)
C:/workdir/g... | NeedsInvestigation | low | Minor |
1,223,256,967 | flutter | Support larger color encodings | Currently the framework uses the `Color` class from dart:ui which expresses all colors as a 32 bit integer.
Problems:
- Flutter may not be able to express all of the colors supported by newer hardware with more than 8 bits of color data per channel.
- Rounding errors from ints to floats throughout color lerpin... | framework,engine,P3,team-engine,triaged-engine | low | Critical |
1,223,278,548 | go | runtime,cmd/link: "found pointer to free object" during asyncPreempt2 on ios-arm64-corellium | ```
##### Test execution environment.
# GOARCH: arm64
# CPU:
# GOOS: ios
# OS Version: Darwin 20.6.0 Darwin Kernel Version 20.6.0: Tue Aug 24 21:10:43 PDT 2021; root:xnu-7195.140.44~1/RELEASE_ARM64_T8010 iPhone9,1
…
# crypto/sha256.test
runtime: marked free object in span 0x100ebd108, elemsize=24 freeindex=8... | OS-Darwin,NeedsInvestigation,mobile,arch-arm64,compiler/runtime | low | Critical |
1,223,281,240 | create-react-app | "process" (and all env vars) not available after upgrade from 4.0.3 to 5.0.1 | ### Describe the bug
After upgrading the `react-scripts` package from 4.0.3 to 5.0.1, trying to run locally only results in a blank white screen in the browser. There is also an error message in the browser console: `Uncaught ReferenceError: process is not defined`. It appears that attempting to reference `process` ... | needs triage,issue: bug report | low | Critical |
1,223,319,602 | pytorch | Performance bad on ARM AArch64 for PyTorch C++ | ### 🚀 The feature, motivation and pitch
Hi PyTorch community, we have been trying to build the PyTorch on the CentOS-7/ Amazon Linux 2 on an aarch64 instance (Graviton) for our C++ project. However, we found our build performance is much worse comparing to the PyTorch aarch64 pip release. I am suspecting if there i... | module: build,triaged,module: arm | low | Major |
1,223,324,309 | deno | Add LSP memory usage to the benchmarks | It would be useful to track the LSP's memory usage. We probably would need to do a bunch of actions and then keep watching memory usage until it stops deviating so much then use that as the reading. | benchmarks,chore,lsp | low | Minor |
1,223,472,928 | pytorch | Torch `x += y.bmm(z)` is faster than `x.baddbmm_(y, z)` | ### 🐛 Describe the bug
The `x.badbmm_(y, z)` method feels like something that should be fast, since it doesn't have to create a temporary tensor for `y.bmm(z)` before it adds it to `x`.
However, from profiling it seems like `badbmm_` just makes a lot of calls to `aten::addmm_` and `aten::select`, and turns out to ... | module: performance,triaged,module: linear algebra | low | Critical |
1,223,480,279 | angular | Type checking for host bindings | Angular does not currently perform type checking for `@Component` and `@Directive` host bindings. This is a gap in Angular's otherwise robust template type checking system that leaves the door open to typos and type mismatches.
- [ ] Introduce an option for host binding type checking
- [ ] Devise and execute a rol... | feature,area: compiler,core: host and host bindings,compiler: template type-checking | low | Minor |
1,223,504,907 | godot | iOS keyboard_input_view confused by arrow keys | ### Godot version
3.4.4
### System information
iOS 15.4.1
### Issue description
Cursor keys, only available with an external keyboard, move the cursor in the hidden GodotKeyboardInputView, but not in the TextEdit, causing the two to get out of sync.
### Steps to reproduce
Connect external keyboard (bluetooth, ... | bug,platform:ios,topic:porting | low | Minor |
1,223,506,673 | pytorch | `reshape` for distributions. | ### 🚀 The feature, motivation and pitch
The `expand` method for distribution instances is great for prepending batch dimensions and, possibly, changing the number of batch elements. Adding similar functionality that mirrors `reshape` on tensors would be useful for manipulating distributions.
For example, suppose ... | module: distributions,feature,triaged | low | Critical |
1,223,507,390 | flutter | Running "flutter create -t module" prints "Found saved certificate choice" and Apple cert info twice | ```
$ flutter create -t module test_module
Found saved certificate choice "Apple Development: Jenn (xxxx)". To clear, use "flutter config".
Signing iOS app for device deployment using developer identity: "Apple Development: Jenn (xxxx)"
Creating project test_module...
Running "flutter pub get" in test_module... ... | platform-ios,tool,P3,team-ios,triaged-ios | low | Minor |
1,223,518,096 | three.js | Suggestion: Rename "Source" to something like "ImageSource" or "TextureSource" | <!-- Ignoring this template may result in your feature request getting deleted -->
**Is your feature request related to a problem? Please describe.**
It's not really clear what "Source" is for from the name.
**Describe the solution you'd like**
I think the class definition could be renamed to be more clear.... | Suggestion | low | Minor |
1,223,566,844 | material-ui | Migrate away from Lerna | ### Duplicates
- [X] I have searched the existing issues
### Latest version
- [X] I have tested the latest version
### Summary 💡
Lerna has been officially marked as "no longer maintained". https://github.com/lerna/lerna/issues/2703 There are definitely options out there, including some approved forks of Lerna. In... | core | low | Minor |
1,223,686,860 | godot | use_mingw="yes" not Honored in custom.py | ### Godot version
4.0.dev (0275d60)
### System information
Windows 10, MinGW, LLVM Build
### Issue description
If I run `scons` in the shell without `use_mingw=yes` even when it's in my `custom.py` file, the build crashes. This seems to be due to the fact that the `if` statement in the `SConstruct` file ... | bug,platform:windows,topic:buildsystem,confirmed | low | Critical |
1,223,741,141 | vscode | Caps Lock Activated Indication in message box that requires a password | <!-- ⚠️⚠️ Do Not Delete This! feature_request_template ⚠️⚠️ -->
<!-- Please read our Rules of Conduct: https://opensource.microsoft.com/codeofconduct/ -->
<!-- Please search existing issues to avoid creating duplicates. -->
<!-- Describe the feature you'd like. -->
Create a "Caps Lock Activated" - Indication when... | feature-request,quick-pick | low | Minor |
1,223,762,843 | vscode | Parenthesis and brace matching within RegExps | <!-- ⚠️⚠️ Do Not Delete This! feature_request_template ⚠️⚠️ -->
<!-- Please read our Rules of Conduct: https://opensource.microsoft.com/codeofconduct/ -->
<!-- Please search existing issues to avoid creating duplicates. -->
<!-- Describe the feature you'd like. -->
Modifying regular expressions is a pain, and a p... | feature-request,editor-bracket-matching | medium | Major |
1,223,841,374 | excalidraw | Highlight words in a paragraph | Hello I don't know if I should drop this idea here but I really wish this feature would become true.
Currently we can't highlight words in a paragraph.
That can be really annoying because it keeps me from making my text more readable.
There are some work arounds with some dividers for example but it's not ideal in... | enhancement,richtext,Candidate P1 | high | Critical |
1,223,894,402 | PowerToys | PowerToys Run - Unit converter - units naming in plural | ### Description of the new feature / enhancement
It would be great that the unit converter can be inputed with mil**s**/meter**s**/mile**s** instead of only mil/mile.
It really is nothing but it would make a bit more sense :)
### Scenario when this would be used?
You say 10 mil**s** and not 10 mil, same for mile**s... | Idea-Enhancement,Product-PowerToys Run,Needs-Triage,Run-Plugin | low | Minor |
1,224,065,080 | pytorch | [JIT] magic methods do not work after reloading model | ### 🐛 Describe the bug
```python
import torch
from torch import Tensor, jit, nn
class Foo(nn.Module):
weight: Tensor
def __init__(self, m: int, n: int):
super().__init__()
self.weight = nn.Parameter(torch.randn(m, n))
@jit.export
def __len__(self) -> int:
return... | oncall: jit | low | Critical |
1,224,074,275 | vscode | Meta Mouse Cursor Mode | This feature request describes a new concept of the "meta mouse cursor mode".
In this mode, the mouse cursor doesn't interact with the UI directly anymore, but shows information about the hovered/clicked UI element and offers options to change its appearance.
The meta-mode intends to provide an intuitive and unif... | feature-request,under-discussion | low | Major |
1,224,132,052 | pytorch | Adding Polyloss to `torch` | ### 🚀 The feature, motivation and pitch
# POLYLOSS: A POLYNOMIAL EXPANSION PERSPEC- TIVE OF CLASSIFICATION LOSS FUNCTIONS
(Published as a conference paper at ICLR 2022) [link to paper](https://arxiv.org/abs/2204.12511)
This loss function implementation won't be difficult as we can use builtin crossentropy loss ... | module: nn,module: loss,triaged | low | Major |
1,224,144,272 | pytorch | test_python_reference_meta_functions takes too long to run | ### 🐛 Describe the bug
Currently, these tests take nearly two minutes to run on CPU:
```
$ python test/test_ops.py -k test_python_reference_meta_functions
................................................... | module: tests,triaged | low | Critical |
1,224,145,579 | material-ui | [migration] How can I replace makeStyles in my code? | I'm sorry, that's kind of a novice question.
I'm currently transferring from material-ui v4 to mui v5 and I used to have a lot of makeStyles and jss styling in the code.
I would like to ask what is the most appropriate way to replace makeStyles for creating classes based on theme variable values. I have code like th... | docs,support: question,v5.x migration | low | Major |
1,224,148,838 | vscode | screenreader reading of inline errors | Issue Type: <b>Feature Request</b>
# feature description
We already have audio queues for when the focus goes over a line that contains errors, perhaps to give us the same experience sighted people have, where they see the errors underlined in red I believe. Next step, the same way sighted people can mouse over t... | feature-request,accessibility,error-list,languages-diagnostics | low | Critical |
1,224,156,981 | pytorch | __torch_function__ callers should always pass kwargs | ### 🐛 Describe the bug
It is such a pain to always have to write:
```
if not kwargs:
kwargs = {}
```
we should just unconditionally pass in the kwargs argument with a fresh empty dictionary
### Versions
master
cc @hameerabbasi @rgommers @peterbell10 | triaged,module: __torch_function__ | low | Critical |
1,224,178,186 | TypeScript | Inconsistent string auto-complete when inside/outside string | # Bug Report
<!--
Please fill in each section completely. Thank you!
-->
### 🔎 Search Terms
<!--
What search terms did you use when trying to find an existing bug report?
List them here so people in the future can find this one more easily.
-->
inconsistent string auto complete intellisense sugg... | Bug | low | Critical |
1,224,188,479 | animate.css | Improve Animations | ## Back Entrances and Exits
We could make them smoother and improve its final steps. They look a bit stiff and shy at the moment.
## Jack in the Box
It could be a bit more "exaggerated".
## Zoom outs
Could be a bit clearer and be more "exaggerated" too.
## Slide Entrances and Exits
Could use more o... | improvement,v5 | low | Minor |
1,224,195,365 | animate.css | Create an easy to use and share playground | Right now it's a bit of an effort to test out all of the animations and work on them. It's 2022, we have better options, let's improve this and make our lives and collaboration easier.
A few things to keep in mind:
- It should be easy to use
- It should have a few different "shapes" for testing out the animations.... | improvement,help wanted,v5 | low | Minor |
1,224,208,437 | material-ui | [Table] Low performance when rendering modest size Table | ### Duplicates
- [X] I have searched the existing issues
### Latest version
- [X] I have tested the latest version
### Current behavior 😯
I'm writing a component which is basically a selectable grid, implemented using `Table`. User can select the cells in a row by pressing and dragging the mouse over th... | performance,component: table | low | Major |
1,224,233,581 | neovim | Lua stdlib feature parity against Vimscript | Tracking issue to collect the current status of Lua stdlib against Vimscript.
As all functions listed here are accessible in Lua via `vim.fn.*`, the goal is **not** to achieve 100% parity, however the typval conversion may have perf concerns and/or functional issues for specific functions so a Lua equivalent may be ... | lua | medium | Critical |
1,224,242,787 | go | runtime: check use of osyield in signal handlers | The SIGPROF handler uses `runtime.osyield` to spin while waiting for exclusive access to the profiling buffer. In the discussion of https://go.dev/cl/400795, @mknyszek questioned whether that was safe. I'm not sure that it is.
I see that on many OSes (Linux, FreeBSD, NetBSD, OpenBSD, Solaris, Dragonfly) that functio... | NeedsInvestigation,compiler/runtime | low | Major |
1,224,246,976 | vscode | Cmd + K, Cmd + I should ideally trigger all hovers | `Cmd + K, Cmd + I` already triggers the editor hover.
For consistency it should also trigger the custom hover in the workbench. This command should also pass the focus to the custom hover for accessibility reasons. | feature-request,accessibility,workbench-hover | low | Minor |
1,224,268,714 | go | runtime: improve the "best effort" of the execution tracer's CPU sample delivery | The work in https://go.dev/cl/400795 (proposed for the Go 1.19 cycle) to add CPU profile samples to the execution tracer's output relies on the process generating at least a modest stream of other traceable events. If the process's goroutines as a group use more than 19 thread-seconds of CPU time before any individual ... | NeedsInvestigation,compiler/runtime | low | Minor |
1,224,271,329 | go | net/http: TestRequestLimit_h2 failure with `didn't expect to get request in Handler` | ```
--- FAIL: TestRequestLimit_h2 (0.48s)
serve_test.go:2969: didn't expect to get request in Handler
FAIL
FAIL net/http 11.329s
```
`greplogs -l -e 'didn'\''t expect to get request in Handler'`
[2022-05-02T14:54:05-349cc83/solaris-amd64-oraclerel](https://build.golang.org/log/9ef0677aafd851270d0fa41479b1d... | NeedsInvestigation | low | Critical |
1,224,278,045 | excalidraw | Disable updating locked elements | When an element is locked, I think we should disable any kind of update (eg changing background, stroke, etc). Right now we disable interaction (eg moving, resizing) but we should also disable any kind of properties update as the element is locked.
Figjam, Miro, whimisical as well don't allow updating elements unles... | discussion,element locking | low | Minor |
1,224,289,348 | go | x/build: OOM while compiling on solaris-amd64-oraclerel | `greplogs --dashboard -md -l -e '\Asolaris-amd64(.*\n)*# .*\nfatal error: .*(?:out of memory|cannot allocate memory)' --since=2022-01-01`
[2022-05-02T20:18:15-8375b54/solaris-amd64-oraclerel](https://build.golang.org/log/c669d551ce9f1bbb0b6d4b9b36742610e88930e3)
```
# net/http_test [net/http.test]
fatal error: runt... | OS-Solaris,Builders,NeedsInvestigation | low | Critical |
1,224,298,814 | excalidraw | Enable selection of single elements to unlock when locked for better UX | As per UX, I feel it's not intuitive to disable selection and I am not sure why we are disabling selection, even when you right-click the element it gets selected.
As per the flow, I select an element -> right-click, -> lock, hence I am expecting then when I click on an element again it will select it, and then I can ... | discussion,element locking | low | Critical |
1,224,307,259 | go | runtime,cmd/compile: SIGSEGV on netbsd-arm64-bsiegert during bootstrapping | `greplogs -l -e '/compile: signal: segmentation fault' --since=2022-01-01`
[2022-05-02T16:21:47-eb4f295-0fdca72/netbsd-arm64-bsiegert](https://build.golang.org/log/268ae7a471439bc52564bd984de09f68154aa178)
[2022-01-06T19:06:36-5b0dc2d-378766a/netbsd-arm64-bsiegert](https://build.golang.org/log/002bd6993330c869cdbcbe5... | OS-NetBSD,NeedsInvestigation,arch-arm64,compiler/runtime | low | Minor |
1,224,309,179 | pytorch | Supporting torch.tensor.apply_ over GPU | ### 🚀 The feature, motivation and pitch
I came across a problem in my project where I needed that `apply_` function. It is interesting to see that it isn't supported over GPU.
I was wondering if it is possible to extend it to GPU, perhaps for pure lambdas to begin with.
If there are no known issues directly s... | module: cuda,triaged | low | Minor |
1,224,336,041 | pytorch | [checkpoint] Handle overlapping storage during save and load | ### 🚀 The feature, motivation and pitch
When loading a checkpoint, we should check whether any of the tensors have overlapping storage and, if so, we're loading data from different storage_keys. We should eliminate the superfluous load is such cases.
A similar check should be performed when saving a checkpoint and... | oncall: distributed,triaged,sharded_tensor | low | Minor |
1,224,346,661 | pytorch | [checkpoint] Stable file format for checkpoints | ### 🚀 The feature, motivation and pitch
We need to provide a stable file format at some level for checkpoints so users can, at the very least, query which PyTorch version is necessary to correctly load it.
### Alternatives
_No response_
### Additional context
_No response_
cc @pietern @mrshenli @pritamdamani... | oncall: distributed,triaged,sharded_tensor | low | Minor |
1,224,357,651 | pytorch | Bug in dataloader iterator found by mypy | While upgrading mypy, found a call to `_BaseDataLoaderIter.reset()` that appears to trivially fail ([link](https://github.com/pytorch/pytorch/blob/543eaac415bfba59e648a3da8be5c4f964f9bc6e/torch/utils/data/dataloader.py#L557)), as it does not provide a required positional argument.
cc @SsnL @VitalyFedyunin @ejguan @Niv... | module: dataloader,triaged | low | Critical |
1,224,373,790 | pytorch | Default repr of __get__ methods in __torch_function__ is bad | ### 🐛 Describe the bug
I got a function in `__torch_function__` that I couldn't handle. I wanted to print it out to see what was up:
```
>>> repr(func)
"<method-wrapper '__get__' of getset_descriptor object at 0x7f1208602a40>"
```
Completely useless. In this case, I needed to call `repr(func.__self__)` to ge... | triaged,module: __torch_function__ | low | Critical |
1,224,406,416 | PowerToys | Instant translator | ### Description of the new feature / enhancement
Given that when selecting any text anywhere, web, desktop, work, excel at the end of the selection then translate to a configured language of PowerToys.
### Scenario when this would be used?
Any location whether on the web or desktop application when text is selected.... | Idea-New PowerToy | low | Major |
1,224,417,295 | pytorch | Quantization in Libtorch | ### 🚀 The feature, motivation and pitch
I am working on a Libtorch program that is heavily performance constrained, and I was hoping to run inference using quantization to speed up the model and reduce memory usage. I would like for Libtorch to support static and dynamic quantization just like Pytorch, both for cre... | oncall: quantization,low priority,triaged | low | Critical |
1,224,422,026 | flutter | Deprecate and remove synthetic `package:flutter_gen` | Currently, `flutter_gen` is a virtual package we generate to provide certain functionality (e.g. [internationalisation](https://docs.flutter.dev/development/accessibility-and-localization/internationalization#adding-your-own-localized-messages)). Developers have to import that package into their app. The package itself... | c: new feature,tool,a: internationalization,P1,team-tool,triaged-tool | medium | Critical |
1,224,423,417 | electron | [Feature Request]: Window always-on-bottom (inverse of top-most) | ### Preflight Checklist
- [X] I have read the [Contributing Guidelines](https://github.com/electron/electron/blob/main/CONTRIBUTING.md) for this project.
- [X] I agree to follow the [Code of Conduct](https://github.com/electron/electron/blob/main/CODE_OF_CONDUCT.md) that this project adheres to.
- [X] I have searc... | enhancement :sparkles:,component/BrowserWindow | low | Minor |
1,224,458,691 | pytorch | hanging process with init_process_group(backend='mpi') cannot be killed | ## Issue description
Running mpirun with torch.distributed hangs. When parts of ranks raised error intentionally and the others are doing allgather, those who raised error cannot be killed and were waiting for all other processes to exit. On the other hand, the process that is doing allgather seems to be waiting for... | oncall: distributed | low | Critical |
1,224,589,613 | flutter | Composing region does not extend while typing on Android emulator | ### Issue
While typing in a Flutter `TextField` on an Android emulator using the software keyboard, when adding to the front of a pre-existing word, the composing region only includes the newly added text. On Android, the composing region extends to include both the newly added text and the rest of the pre-existing wo... | a: text input,platform-android,framework,P2,team-android,triaged-android | low | Minor |
1,224,644,296 | flutter | Make `Theme` a UbiquitousInheritedWidget | In https://github.com/flutter/flutter/pull/102336 we introduced the idea of the `_UbiquitousInheritedWidget` for `Directionality` which makes a tradeoff between faster build performance versus worse update performance. I think we should consider moving `Theme` over to `UbiquitousInheritedWidget` as well. All my other... | framework,f: material design,c: proposal,team-design,triaged-design | low | Major |
1,224,665,507 | flutter | [flutter_tools] RPCError at new _OutstandingRequest at package:vm_service/src/vm_service.dart | on `beta/2.13.0-0.3.pre`, with command `flutter run -d chrome` on Windows 10
```
Thread 0 main threadRPCError: ext.flutter.debugDumpApp: (-32603) ext.flutter.debugDumpApp: Unexpected error from chrome devtools:text: Uncaught (in promise)exception: description: null type: string value: Unexpected null value.attem... | c: crash,tool,platform-web,P2,team-web,triaged-web | low | Critical |
1,224,672,657 | neovim | Text getter/setter using character columns as the horizontal coordinate | ### Feature already in Vim?
N/A
### Feature description
As a plugin author, I often draw text into random-access locations in a buffer to create a TUI. Right now, the functions `nvim_buf_[get|set]_text` use byte offsets for the horizontal coordinates. As a result, if I want to put text into a certain character... | enhancement,api | low | Minor |
1,224,740,512 | pytorch | Support Positional-only Arguments in JIT | ### 🚀 The feature, motivation and pitch
From Python 3.8, we have positional-only arguments for function definition which gives a better control of code behaviors.
However, it seems that PyTorch (1.11) JIT (`torch.jit.script`) does not support it.
For instance, if we replace the example code
```python
import torch... | oncall: jit | low | Critical |
1,224,741,983 | go | x/tools/gopls: eagerly compute the workspace symbol index | This is a spin-off of #52602, which we're going to close as just an optiimization.
In the future, we should go one step further and populate the workspace symbol index at load time. There's a straightforward way to do this: while type-checking we _always_ have access to the full AST of each package, before filtering... | gopls,Tools,gopls/performance | low | Major |
1,224,754,359 | material-ui | [IconButton][ClickAwayListener] Unexpected behavior depending on where button is clicked | ### Duplicates
- [X] I have searched the existing issues
### Latest version
- [X] I have tested the latest version
### Current behavior 😯
When the child of a ClickAwayListener component is visible, and the user clicks on the SVG element (light blue in example) of an IconButton, the child hides as expect... | bug 🐛,component: icon button,component: ClickAwayListener | low | Critical |
1,224,769,749 | pytorch | [JIT] Infinite RecursionError with self-referential models (also affects `__repr__`)!! | ### 🐛 Describe the bug
So, this may be a bit unorthodox, but I am working with invertible models, i.e. models for (left/right) invertible functions. These are `nn.Module`s that not only have a `.forward`, but also a `.inverse`. A popular example would be the [invertible Residual Network](https://proceedings.mlr.pre... | oncall: jit | low | Critical |
1,224,783,481 | youtube-dl | Pls support UDel free livestreams | <!--
######################################################################
WARNING!
IGNORING THE FOLLOWING TEMPLATE WILL RESULT IN ISSUE CLOSED AS INCOMPLETE
######################################################################
-->
## Checklist
<!--
Carefully read and work through this check lis... | site-support-request | low | Critical |
1,224,808,462 | pytorch | `torch.addmv` backward fails | ### 🐛 Describe the bug
`torch.addmv` backward fails when `input` is `float64`, `mat` and `vec` are `complex128`
```python
import torch
input = torch.rand([2], dtype=torch.float64, requires_grad=True)
mat = torch.rand([2, 3], dtype=torch.complex128, requires_grad=True)
vec = torch.rand([3], dtype=torch.complex1... | module: autograd,triaged,module: complex,actionable | low | Critical |
1,224,813,963 | pytorch | Strange warning from `matmul(..., out=...)` | ### 🐛 Describe the bug
The torch.matmul `out=` parameter seems to reshape its parameter down to two dimensions, which leads to a strange UserWarning:
```
U = torch.randn(2, 4, 3)
W = torch.randn(3, 5)
temp = torch.empty((U @ W).shape)
torch.matmul(U, W, out=temp)
UserWarning: An output with one or more elements... | triaged,module: linear algebra | low | Critical |
1,224,824,193 | pytorch | `gradcheck` should support the comparison of NaN | ### 🐛 Describe the bug
`gradcheck` should support the comparison of NaN, like adding an argument that whether `NaN == NaN` in the check.
```python
import torch
input = torch.rand([4], dtype=torch.float64, requires_grad=True)
torch.autograd.gradcheck(torch.acosh, (input))
# torch.autograd.gradcheck.GradcheckE... | module: autograd,triaged,enhancement,actionable | low | Critical |
1,224,849,440 | flutter | [Proposal] Add `retry` to `testWidgets` | Hi thanks for the framework! I notice that `testWidgets` has no `retry` flag while `test` has. Therefore, it is impossible to retry a flaky widget test (or integration test) while it is possible for a unit test.
The solution may be quite simple: Just forward that argument. | a: tests,c: new feature,framework,c: proposal,P3,team-framework,triaged-framework | low | Major |
1,224,852,623 | pytorch | `gradcheck` fails for `torch.trace` | ### 🐛 Describe the bug
`gradcheck` fails for `torch.trace` when `dim1 > dim0 + 1`
```python
import torch
input = torch.rand([5, 3], dtype=torch.float64, requires_grad=True)
torch.autograd.gradcheck(torch.trace, (input))
# torch.autograd.gradcheck.GradcheckError: Jacobian mismatch for output 0 with respect ... | module: autograd,triaged,actionable | low | Critical |
1,224,860,544 | flutter | PrerollContext.has_texture_layer is sticky and impacts layers unrelated to the TextureLayer | Looking at the code in ContainerLayer::PrerollChildren, there are 2 flags that are accumulated on a subtree basis: has_platform_layer and has_texture_layer. The has_platform_layer flag appears to be accumulated within a subtree but only affects parents and ancestors. The has_texture_layer flag, however, is only ever se... | engine,waiting for PR to land (fixed),P2,team-engine,triaged-engine | low | Minor |
1,224,864,900 | pytorch | `addr, baddmm, dist, l1_loss` will backward fail when input tensors have different dtypes | ### 🐛 Describe the bug
`addr, baddmm, dist, l1_loss` will backward fail when input tensors have different dtypes
```python
import torch
input = torch.rand([3, 2], dtype=torch.float64, requires_grad=True)
vec1 = torch.rand([3], dtype=torch.float64, requires_grad=True)
vec2 = torch.rand([2], dtype=torch.complex1... | module: autograd,triaged,module: complex,actionable,complex_autograd | low | Critical |
1,224,866,374 | pytorch | `cumprod, prod` will backward fail if `dtype` argument is different than the dtype of input tensor | ### 🐛 Describe the bug
`cumprod, prod` will backward fail if `dtype` argument is different than the dtype of input tensor
```python
import torch
def fn(input):
dim = -1
dtype = torch.complex128
fn_res = torch.cumprod(input, dim, dtype=dtype)
return fn_res
input = torch.rand([5, 5, 5], dtyp... | module: autograd,triaged,module: complex,complex_autograd | low | Critical |
1,224,875,820 | yt-dlp | [Fifa] Some videos generate 404 error | ### Checklist
- [X] I'm reporting a broken site
- [X] I've verified that I'm running yt-dlp version **2022.04.08** ([update instructions](https://github.com/yt-dlp/yt-dlp#update)) or later (specify commit)
- [X] I've checked that all provided URLs are alive and playable in a browser
- [X] I've checked that all URLs an... | DRM,site-enhancement,site-bug | low | Critical |
1,224,886,291 | create-react-app | Looking forward to official support for no eject extension react-scripts/webpack | ### Extended configuration is required
When we create react app with create-react-app, we almost always need to extend more configurations ( react-scripts or webpack ), and we want to not eject to continuously upgrade react-scripts.
The current available methods are as follows, but they all have their own problem... | issue: proposal,needs triage | medium | Major |
1,224,937,731 | youtube-dl | feature request - trailer of youtube premiere | <!--
######################################################################
WARNING!
IGNORING THE FOLLOWING TEMPLATE WILL RESULT IN ISSUE CLOSED AS INCOMPLETE
######################################################################
-->
## Checklist
<!--
Carefully read and work through this check lis... | broken-IE | low | Critical |
1,224,957,254 | PowerToys | MKV file thumbnail preview, artwork preview. | ### Description of the new feature / enhancement
MKV file thumbnail preview, artwork preview.
### Scenario when this would be used?
MKV file thumbnail preview, artwork preview.
### Supporting information
MKV file thumbnail preview, artwork preview windows does not support this feature. | Idea-Enhancement,Product-File Explorer,Needs-Triage | low | Minor |
1,224,961,635 | godot | [Mono] 64-bit real numbers (`Variant::REAL`) converted to `System.Single` in C# | ### Godot version
3.4.4 stable mono c#
### System information
win10 gles3 gtx1050ti
### Issue description
c# JSON.parse display and precision errors
### Steps to reproduce
) or later (specify commit)
- [X] I've checked that all provided URLs are alive and playable in a browser
- [X] I've checked that all UR... | site-bug,patch-available | low | Critical |
1,225,024,508 | godot | screen reading shader samples wrong color if unlit sprite is in front of sprite lit by a Light2D | ### Godot version
v3.4.4.stable.official [419e713a2]
### System information
Windows 10, Nvidia Geforce 2060
### Issue description
One sprite is in the background and is lit by a light2d node. I have an u... | bug,topic:rendering,topic:shaders,topic:2d | low | Critical |
1,225,025,670 | flutter | [go_router_builder] Make kebab-case optional for generated query parameters | <!-- Thank you for using Flutter!
If you are looking for support, please check out our documentation
or consider asking a question on Stack Overflow:
* https://flutter.dev/
* https://api.flutter.dev/
* https://stackoverflow.com/questions/tagged/flutter?sort=frequent
If you hav... | c: new feature,package,c: proposal,P3,p: go_router_builder,team-go_router,triaged-go_router | low | Critical |
1,225,058,625 | pytorch | `gradcheck` for `torch.solve` may trigger INTERNAL ASSERT FAILED | ### 🐛 Describe the bug
`gradcheck` for `torch.solve` may trigger INTERNAL ASSERT FAILED randomly.
```python
import torch
input = torch.rand([0, 2, 5, 4], dtype=torch.float64, requires_grad=True)
A = torch.rand([1, 5, 5], dtype=torch.float64, requires_grad=True)
torch.autograd.gradcheck(torch.solve, (input,... | needs reproduction,module: autograd,triaged | low | Critical |
Subsets and Splits
GitHub Issues Containing Next.js References
Filters training data to find examples mentioning "next.js", providing basic keyword search capability but offering limited analytical value beyond simple retrieval.