starred/terminal
1
0
Fork 0
mirror of https://github.com/microsoft/terminal.git synced 2026-02-03 21:25:34 +00:00
Code Issues 19.0k Packages Projects Releases 20 Wiki Activity

Better signaling between processes #3514

New Issue
Closed
opened 2026-01-30 23:22:56 +00:00 by claunia · 3 comments
claunia commented 2026-01-30 23:22:56 +00:00
Owner
Copy Link

Originally created by @giggio on GitHub (Aug 21, 2019).

Originally assigned to: @bitcrazed on GitHub.

Description of the new feature/enhancement

Provide better signal passing between processes, something similar to what Linux has to offer.

(I am aware this is not perhaps the best place to start this conversation, if you can find a better home for this issue, I'll gladly move it there.)

On Linux/Unix a process can signal another process very easily, and the process receiving the signal can also, easily trap it and decide how to react. So, ctrl+c, por example, generates a SIGINT signal, and the process receiving that signal can decide how to answer to that signal. In bash, for example, it is really easy to trap those signals and respond to them. Processes can also send signals other ways, like using kill.

On Windows it is really hard to work with "signals", even with ctrl+c, and the "signals" you can send are limited. Also, during my investigations, I noticed that, by default, if a parent process gets a ctrl+c, all it's childs also will, and if the parent disables that and spawns childs, and then reenable, it is then impossible to signal the children later.

I am trying to work with a console app that takes signal, and decides which processes it will forward the signal to. It doesn't seem possible on Windows. On *NIX it would be trivial.

Proposed technical implementation details (optional)

I am aware that this team is really careful to not break legacy code, so what I propose is a new signal layer, while keeping the old one untouched. This new layer could forward signals to the old one, and also get signals from the old one, while keeping compatibility. I am not aware how to implement that in a way that does not break things.

Originally created by @giggio on GitHub (Aug 21, 2019). Originally assigned to: @bitcrazed on GitHub. # Description of the new feature/enhancement Provide better signal passing between processes, something similar to what Linux has to offer. (I am aware this is not perhaps the best place to start this conversation, if you can find a better home for this issue, I'll gladly move it there.) On Linux/Unix a process can signal another process very easily, and the process receiving the signal can also, easily trap it and decide how to react. So, <kbd>ctrl</kbd>+<kbd>c</kbd>, por example, generates a `SIGINT` signal, and the process receiving that signal can decide how to answer to that signal. In bash, for example, it is really easy to trap those signals and respond to them. Processes can also send signals other ways, like using `kill`. On Windows it is really hard to work with "signals", even with <kbd>ctrl</kbd>+<kbd>c</kbd>, and the "signals" you can send are limited. Also, during my investigations, I noticed that, by default, if a parent process gets a <kbd>ctrl</kbd>+<kbd>c</kbd>, all it's childs also will, and if the parent disables that and spawns childs, and then reenable, it is then impossible to signal the children later. I am trying to work with a console app that takes signal, and decides which processes it will forward the signal to. It doesn't seem possible on Windows. On *NIX it would be trivial. # Proposed technical implementation details (optional) I am aware that this team is really careful to not break legacy code, so what I propose is a new signal layer, while keeping the old one untouched. This new layer could forward signals to the old one, and also get signals from the old one, while keeping compatibility. I am not aware how to implement that in a way that does not break things.
claunia added the Issue-FeatureNeeds-Tag-FixResolution-Won't-Fix labels 2026-01-30 23:22:57 +00:00
claunia commented 2026-01-30 23:22:57 +00:00
Author
Owner
Copy Link

@DHowett-MSFT commented on GitHub (Aug 22, 2019):

handing off to @bitcrazed 😄

@DHowett-MSFT commented on GitHub (Aug 22, 2019): handing off to @bitcrazed :smile:
claunia commented 2026-01-30 23:22:57 +00:00
Author
Owner
Copy Link

@DHowett-MSFT commented on GitHub (Aug 22, 2019):

This may be quite related to #281 (which is like SIGWINCH)

@DHowett-MSFT commented on GitHub (Aug 22, 2019): This may be quite related to #281 (which is like `SIGWINCH`)
claunia commented 2026-01-30 23:22:58 +00:00
Author
Owner
Copy Link

@bitcrazed commented on GitHub (Aug 29, 2019):

Hey @giggio Appreciate the ask.

However, this is not something we're going to be able to support.

*NIX and Windows follow very different philosophical and architectural paths.

In *NIX, signals are used for many inter-process communications (IPC), but Windows has several IPC mechanisms that are designed for various specific scenarios, including RPC, pipes, mailslots, etc., and several interprocess synchronization and other synchronization mechanisms (e.g. locks, semaphores, mutexes, timers, and events), along with several thread/process control mechanisms. And because these are all kernel-provided mechanisms, they can all be ACLed, and managed in sophisticated ways.

These mechanisms delivers to Windows developers a great deal of power, flexibility, and control - moreso, perhaps that *NIX offers - but ... as you point out ... there's no direct 1:1 mapping with *NIX signals.

Could this be remedied? Maybe. Possibly. But, alas, it's way out of our sphere of influence and control. However, it is something that we hear from time to time, and we do on occasion do what we can in the command-line to better supported expected command-line scenarios, but WE aren't going to be able to inject a signalling mechanism into the kernel ;)

Closing since there's no further action we can perform to make this happen . Sorry!

@bitcrazed commented on GitHub (Aug 29, 2019): Hey @giggio Appreciate the ask. However, this is not something we're going to be able to support. *NIX and Windows follow very different philosophical and architectural paths. In *NIX, signals are used for many inter-process communications (IPC), but [Windows has several IPC mechanisms](https://docs.microsoft.com/en-us/windows/win32/ipc/interprocess-communications) that are designed for various specific scenarios, including RPC, pipes, mailslots, etc., and several [interprocess synchronization](https://docs.microsoft.com/en-us/windows/win32/sync/interprocess-synchronization) and [other synchronization](https://docs.microsoft.com/en-us/windows/win32/sync/about-synchronization) mechanisms (e.g. locks, semaphores, mutexes, timers, and events), along with several [thread/process control mechanisms](https://docs.microsoft.com/en-us/windows/win32/procthread/process-and-thread-functions). And because these are all kernel-provided mechanisms, they can all be ACLed, and managed in sophisticated ways. These mechanisms delivers to Windows developers a great deal of power, flexibility, and control - moreso, perhaps that *NIX offers - but ... as you point out ... there's no direct 1:1 mapping with *NIX signals. Could this be remedied? Maybe. Possibly. But, alas, it's way out of our sphere of influence and control. However, it is something that we hear from time to time, and we do on occasion do what we can in the command-line to better supported expected command-line scenarios, but WE aren't going to be able to inject a signalling mechanism into the kernel ;) Closing since there's no further action we can perform to make this happen . Sorry!
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
⛺ Reserved
A11yCO
A11yMAS
A11ySev1
A11ySev2
A11ySev3
A11yTTValidated
A11yUsable
A11yVoiceAccess
A11yWCAG
Area-Accessibility
Area-AtlasEngine
Area-AzureShell
Area-Build
Area-Build
Area-Chat
Area-CmdPal
Area-CodeHealth
Area-Commandline
Area-CookedRead
Area-DefApp
Area-Extensibility
Area-Fonts
Area-GroupPolicy
Area-i18n
Area-Input
Area-Interaction
Area-Interop
Area-Localization
Area-Output
Area-Performance
Area-Portable
Area-Quality
Area-Remoting
Area-Rendering
Area-Schema
Area-Server
Area-Settings
Area-SettingsUI
Area-ShellExtension
Area-ShellExtension
Area-ShellExtension
Area-Suggestions
Area-Suggestions
Area-TerminalConnection
Area-TerminalControl
Area-Theming
Area-UserInterface
Area-VT
Area-Windowing
Area-WPFControl
AutoMerge
Blocking-Ingestion
Culprit-Centennial
Culprit-WinUI
Disability-All
Disability-Blind
Disability-LowVision
Disability-Mobility
External-Blocked-WinUI3
Fixed
Gathering-Data
good first issue
HCL-E+D
HCL-WindowsTerminal
Help Wanted
Impact-Compatibility
Impact-Compliance
Impact-Correctness
Impact-Visual
In-PR
InclusionBacklog
InclusionBacklog-Windows TerminalWin32
InclusionCommitted-202206
Issue-Bug
Issue-Docs
Issue-Feature
Issue-Feature
Issue-Question
Issue-Samples
Issue-Scenario
Issue-Task
Needs-Attention
Needs-Author-Feedback
Needs-Bisect
Needs-Discussion
Needs-Repro
Needs-Tag-Fix
Needs-Tag-Fix
Needs-Triage
No-Recent-Activity
Priority-0
Priority-1
Priority-2
Priority-3
Product-Cmd.exe
Product-Colortool
Product-Colortool
Product-Colortool
Product-Conhost
Product-Conpty
Product-Meta
Product-Powershell
Product-Terminal
Product-WSL
pull-request
Mirrored from GitHub Pull Request
 Resolution-Answered
Resolution-By-Design
Resolution-Duplicate
Resolution-External
Resolution-Fix-Available
Resolution-Fix-Committed
Resolution-No-Repro
Resolution-Won't-Fix
Severity-Blocking
Severity-Crash
Severity-DataLoss
spam
this-will-be-a-breaking-change
Tracking-External
WindowsTerminal_Win32
Work-Item
zAskModeBug
zInbox-Bug
No Label Issue-Feature Needs-Tag-Fix Resolution-Won't-Fix
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
claunia
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: starred/terminal#3514
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?