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

BUG: Minimizing Terminal Crash #2516

New Issue
Closed
opened 2026-01-30 22:57:18 +00:00 by claunia · 6 comments
claunia commented 2026-01-30 22:57:18 +00:00
Owner
Copy Link

Originally created by @evlli on GitHub (Jul 3, 2019).

Originally assigned to: @zadjii-msft on GitHub.

Environment

Windows build number: 10.0.18362.175
Windows Terminal version: 0.2.1831.0

Steps to reproduce

Execute a command that keeps outputting something for a while.
for example you can user apt list or ls in a directory with a lot of content.

Then, while the command is running, double click on the titlebar or click the minimize? button.

Expected behavior

The behavior I would have expected was for it to pause or keep running

Actual behavior

The Terminal Freezes and crashes after approximately 1,5 seconds
Other Terminal windows are unaffected

Originally created by @evlli on GitHub (Jul 3, 2019). Originally assigned to: @zadjii-msft on GitHub. <!-- This bug tracker is monitored by Windows Terminal development team and other technical folks. **Important: When reporting BSODs or security issues, DO NOT attach memory dumps, logs, or traces to Github issues**. Instead, send dumps/traces to secure@microsoft.com, referencing this GitHub issue. Please use this form and describe your issue, concisely but precisely, with as much detail as possible. --> # Environment ```none Windows build number: 10.0.18362.175 Windows Terminal version: 0.2.1831.0 ``` # Steps to reproduce Execute a command that keeps outputting something for a while. for example you can user `apt list` or `ls` in a directory with a lot of content. Then, while the command is running, double click on the titlebar or click the minimize? button. # Expected behavior The behavior I would have expected was for it to pause or keep running # Actual behavior The Terminal Freezes and crashes after approximately 1,5 seconds Other Terminal windows are unaffected
claunia commented 2026-01-30 22:57:19 +00:00
Author
Owner
Copy Link

@zadjii-msft commented on GitHub (Jul 3, 2019):

I'm not sure if it's the same crash, but while trying to repro this, I managed to get an exception while restoring down from maximize while the app was outputting.

...
10 (Inline Function) --------`-------- conhost!wil::details::in1diag3::Throw_HrIf+0x17 [c:\ba\2009\s\dep\wil\include\wil\result_macros.h @ 5039] 
11 0000008b`002ff6d0 00007ff6`ceb4e2e5 conhost!TextBufferCellIterator::TextBufferCellIterator+0x281 [c:\ba\2009\s\src\buffer\out\textbuffercelliterator.cpp @ 43] 
12 (Inline Function) --------`-------- conhost!TextBuffer::GetCellDataAt+0x19 [c:\ba\2009\s\src\buffer\out\textbuffer.cpp @ 169] 
13 0000008b`002ff740 00007ff6`ceb4cefa conhost!Microsoft::Console::Render::Renderer::_PaintBufferOutput+0x1c5 [c:\ba\2009\s\src\renderer\base\renderer.cpp @ 573] 
14 0000008b`002ff8b0 00007ff6`ceb4ccca conhost!Microsoft::Console::Render::Renderer::_PaintFrameForEngine+0x1ea [c:\ba\2009\s\src\renderer\base\renderer.cpp @ 112] 
15 0000008b`002ff9a0 00007ff6`ceb4f6a2 conhost!Microsoft::Console::Render::Renderer::PaintFrame+0x6a [c:\ba\2009\s\src\renderer\base\renderer.cpp @ 65] 
16 (Inline Function) --------`-------- conhost!Microsoft::Console::Render::RenderThread::_ThreadProc+0x40 [c:\ba\2009\s\src\renderer\base\thread.cpp @ 165] 
17 0000008b`002ff9d0 00007ffd`26ef18a4 conhost!Microsoft::Console::Render::RenderThread::s_ThreadProc+0x52

Looked like it was trying to draw one of the lines that's now outside the viewport. Perhaps we're not locking the terminal properly for certain resize types? Or we've moved the viewport to a bad position on restore down? Fortunately, the repro is easy enough, so we just need to dig in more.

Thanks!

@zadjii-msft commented on GitHub (Jul 3, 2019): I'm not sure if it's the same crash, but while trying to repro this, I managed to get an exception while _restoring down_ from maximize while the app was outputting. ``` ... 10 (Inline Function) --------`-------- conhost!wil::details::in1diag3::Throw_HrIf+0x17 [c:\ba\2009\s\dep\wil\include\wil\result_macros.h @ 5039] 11 0000008b`002ff6d0 00007ff6`ceb4e2e5 conhost!TextBufferCellIterator::TextBufferCellIterator+0x281 [c:\ba\2009\s\src\buffer\out\textbuffercelliterator.cpp @ 43] 12 (Inline Function) --------`-------- conhost!TextBuffer::GetCellDataAt+0x19 [c:\ba\2009\s\src\buffer\out\textbuffer.cpp @ 169] 13 0000008b`002ff740 00007ff6`ceb4cefa conhost!Microsoft::Console::Render::Renderer::_PaintBufferOutput+0x1c5 [c:\ba\2009\s\src\renderer\base\renderer.cpp @ 573] 14 0000008b`002ff8b0 00007ff6`ceb4ccca conhost!Microsoft::Console::Render::Renderer::_PaintFrameForEngine+0x1ea [c:\ba\2009\s\src\renderer\base\renderer.cpp @ 112] 15 0000008b`002ff9a0 00007ff6`ceb4f6a2 conhost!Microsoft::Console::Render::Renderer::PaintFrame+0x6a [c:\ba\2009\s\src\renderer\base\renderer.cpp @ 65] 16 (Inline Function) --------`-------- conhost!Microsoft::Console::Render::RenderThread::_ThreadProc+0x40 [c:\ba\2009\s\src\renderer\base\thread.cpp @ 165] 17 0000008b`002ff9d0 00007ffd`26ef18a4 conhost!Microsoft::Console::Render::RenderThread::s_ThreadProc+0x52 ``` Looked like it was trying to draw one of the lines that's now outside the viewport. Perhaps we're not locking the terminal properly for certain resize types? Or we've moved the viewport to a bad position on restore down? Fortunately, the repro is easy enough, so we just need to dig in more. Thanks!
claunia commented 2026-01-30 22:57:19 +00:00
Author
Owner
Copy Link

@zadjii-msft commented on GitHub (Jul 26, 2019):

Leaving notes here for me after the weekend:

Trying to repro this, the terminal moves between [118x26]<->[207x61]

When we crash, the conpty's buffer has 61 rows with 118 chars each. We're trying to render row 60, and we want all 207 characters from it. Seems to imply that the buffer's width wasn't resized, but the height was.

That's not right.

@zadjii-msft commented on GitHub (Jul 26, 2019): Leaving notes here for me after the weekend: Trying to repro this, the terminal moves between [**118**x26]<->[207x**61**] When we crash, the conpty's buffer has **61** rows with **118** chars each. We're trying to render row 60, and we want all **207** characters from it. Seems to imply that the buffer's width wasn't resized, but the height _was_. That's not right.
claunia commented 2026-01-30 22:57:19 +00:00
Author
Owner
Copy Link

@zadjii-msft commented on GitHub (Jul 29, 2019):

Okay debugging more into this. For my repro, the resize is from [207x61] -> [118x26]

When we resize in ConPty, we call both SetConsoleScreenBufferInfoEx and if that succeeds, SetConsoleWindowInfo. In SetConsoleScreenBufferInfoEx, we'll perform 2 ResizeWithReflow operations(!).

  • The first will resize from [207x61] -> [118x61], changing the width of all the rows.
  • The second will resize from [118x61] -> [118x26], changing the height of the buffer. The second resize is triggered by the ScreenInfo::SetViewportSize call near the bottom ofSetConsoleScreenBufferInfoEx.
    • During this resize, we'll copy the first 26 rows (up to iOldRow = 25) just fine. However, the problem comes when we try to newline the cursor. When that happens on the 25th row, we'll attempt to circle the buffer, to shift all the buffer contents up a row, and give us a new row to work with. Hypothetically, this will happen 35 more times, as we copy all the old lines into the new buffer.
    • HOWEVER, as we call TextBuffer::IncrementCircularBuffer, we'll also tell the renderer that it should give the render engines a chance to paint, before any old text was lost. This is helpful when a newline is normally output to the terminal, but isn't really necessary to do during a resize.
    • The renderer will ask the VT engine if it wants to paint to handle the buffer circling. The VT engine will always say "yes".
    • The renderer will then immediately render a new frame for the VT engine. Somewhere during this frame, the vt engine will throw an exception.
    • The exception will get caught way back up in SetConsoleScreenBufferInfoEx. The rest of the buffer will fail to get copied, and even worse, we won't even move the [118x26] text buffer into the screen buffer, leaving the screen buffer with a incorrectly sized buffer.

Later on, we'll try and render a new frame, and try to render the 60th line of the buffer, with a width of 207 chars. The buffer does incorrectly still have 61 lines, but it only has 118 chars per row, so the renderer throws an exception that's not caught here, crashing conpty. Conpty crashing causes the terminal to crash.

Now we just need to come up with the right solution for handling this resize. Very simply, we could maybe tell conpty to not paint at all during a resize. Might be too risky a change for 0.3, however.

@zadjii-msft commented on GitHub (Jul 29, 2019): Okay debugging more into this. For my repro, the resize is from [207x61] -> [118x26] When we resize in ConPty, we call both `SetConsoleScreenBufferInfoEx` and if that succeeds, `SetConsoleWindowInfo`. In `SetConsoleScreenBufferInfoEx`, we'll perform **_2_** `ResizeWithReflow` operations(**!**). * The first will resize from [207x61] -> [118x61], changing the _width_ of all the rows. * The second will resize from [118x61] -> [118x26], changing the height of the buffer. The second resize is triggered by the `ScreenInfo::SetViewportSize` call near the bottom of`SetConsoleScreenBufferInfoEx`. - During this resize, we'll copy the first 26 rows (up to `iOldRow = 25`) just fine. However, the problem comes when we try to newline the cursor. When that happens on the 25th row, we'll attempt to circle the buffer, to shift all the buffer contents up a row, and give us a new row to work with. Hypothetically, this will happen 35 more times, as we copy all the old lines into the new buffer. - HOWEVER, as we call `TextBuffer::IncrementCircularBuffer`, we'll also tell the renderer that it should give the render engines a chance to paint, before any old text was lost. This is helpful when a newline is normally output to the terminal, but isn't really necessary to do during a resize. - The renderer will ask the VT engine if it wants to paint to handle the buffer circling. The VT engine will always say "yes". - The renderer will then _immediately_ render a new frame for the VT engine. **Somewhere during this frame**, the vt engine will throw an exception. - The exception will get caught way back up in `SetConsoleScreenBufferInfoEx`. The rest of the buffer will fail to get copied, and even worse, we _won't even move_ the [118x26] text buffer into the screen buffer, leaving the screen buffer with a incorrectly sized buffer. Later on, we'll try and render a new frame, and try to render the 60th line of the buffer, with a width of 207 chars. The buffer _does_ incorrectly still have 61 lines, but it only has 118 chars per row, so the renderer throws an exception that's _not_ caught here, crashing conpty. Conpty crashing causes the terminal to crash. Now we just need to come up with the right solution for handling this resize. Very simply, we could maybe tell conpty to not paint at all during a resize. Might be too risky a change for 0.3, however.
claunia commented 2026-01-30 22:57:20 +00:00
Author
Owner
Copy Link

@mcpiroman commented on GitHub (Jul 29, 2019):

@zadjii-msft Maybe, kind of, #1946? It got lost somewhere in the swarm of issues few weeks ago :(

@mcpiroman commented on GitHub (Jul 29, 2019): @zadjii-msft Maybe, kind of, #1946? It got lost somewhere in the swarm of issues few weeks ago :(
claunia commented 2026-01-30 22:57:20 +00:00
Author
Owner
Copy Link

@zadjii-msft commented on GitHub (Jul 29, 2019):

@mcpiroman It's probably related to the larger class of "conpty resizing issues". I think it might be able to improve the QOL for the terminal by fixing this particular bug now, while a more complete design of the right way to handle conpty resizing is a while off. I do still owe you a response on that other issue though.

@zadjii-msft commented on GitHub (Jul 29, 2019): @mcpiroman It's probably related to the larger class of "conpty resizing issues". I think it might be able to improve the QOL for the terminal by fixing this particular bug _now_, while a more complete design of the _right_ way to handle conpty resizing is a while off. I do still owe you a response on that other issue though.
claunia commented 2026-01-30 22:57:20 +00:00
Author
Owner
Copy Link

@ghost commented on GitHub (Aug 3, 2019):

:tada:This issue was addressed in #2149, which has now been successfully released as Windows Terminal Preview v0.3.2142.0.🎉

Handy links:

@ghost commented on GitHub (Aug 3, 2019): :tada:This issue was addressed in #2149, which has now been successfully released as `Windows Terminal Preview v0.3.2142.0`.:tada: Handy links: * [Release Notes](https://github.com/microsoft/terminal/releases/tag/v0.3.2142.0) * [Store Download](https://www.microsoft.com/store/apps/9n0dx20hk701?cid=storebadge&ocid=badge)
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 Area-TerminalControl Issue-Bug Needs-Tag-Fix Product-Conpty Resolution-Fix-Committed Severity-Crash
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#2516
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?