starred/terminal
1
0
Fork 0
mirror of https://github.com/microsoft/terminal.git synced 2026-02-04 05:35:20 +00:00
Code Issues 19.0k Packages Projects Releases 20 Wiki Activity

Crash in OnAppInitialized #19989

New Issue
Closed
opened 2026-01-31 06:59:42 +00:00 by claunia · 3 comments
claunia commented 2026-01-31 06:59:42 +00:00
Owner
Copy Link

Originally created by @zadjii-msft on GitHub (May 26, 2023).

Two bugs, pretty sure same root cause. I believe this is collectively 21% of the crashes in 1.18 so far.

  • MSFT:44725712 "WindowsTerminal.exe!NonClientIslandWindow::OnSize"
  • MSFT:44754014 "NonClientIslandWindow::GetTotalNonClientExclusiveSize"

Both have stacks north of WindowsTerminal.exe!AppHost::_WindowInitializedHandler$_ResumeCoro$

3c3b1aac02/src/cascadia/WindowsTerminal/AppHost.cpp (L1208-L1237)

See the bug?

I'm betting that the app exits before the coroutine gets scheduled. That's a hunch.

North of the ShowWindow call, user32 calls back to our WndProc:

L: 35 | 100.00%WindowsTerminal.exe!NonClientIslandWindow::GetTotalNonClientExclusiveSize
  L: 45 | 100.00%WindowsTerminal.exe!IslandWindow::MessageHandler
    L: 55 | 100.00%WindowsTerminal.exe!NonClientIslandWindow::MessageHandler
      L: 65 | 100.00%WindowsTerminal.exe!BaseWindow_IslandWindow_::WndProc

L: 426 | 100.00%WindowsTerminal.exe!NonClientIslandWindow::OnSize
  L: 526 | 100.00%WindowsTerminal.exe!IslandWindow::OnResize
    L: 626 | 100.00%WindowsTerminal.exe!IslandWindow::MessageHandler
      L: 726 | 100.00%WindowsTerminal.exe!NonClientIslandWindow::MessageHandler
        L: 826 | 100.00%WindowsTerminal.exe!BaseWindow_IslandWindow_::WndProc

Loading dumps.

Originally created by @zadjii-msft on GitHub (May 26, 2023). Two bugs, pretty sure same root cause. I believe this is collectively **21% of the crashes** in 1.18 so far. * MSFT:44725712 "WindowsTerminal.exe!NonClientIslandWindow::OnSize" * MSFT:44754014 "NonClientIslandWindow::GetTotalNonClientExclusiveSize" Both have stacks north of `WindowsTerminal.exe!AppHost::_WindowInitializedHandler$_ResumeCoro$` https://github.com/microsoft/terminal/blob/3c3b1aac02320f49f6887f08e6d6d1d56132b1ef/src/cascadia/WindowsTerminal/AppHost.cpp#L1208-L1237 See the bug? I'm betting that the app exits before the coroutine gets scheduled. That's a hunch. North of the `ShowWindow` call, user32 calls back to our WndProc: ``` L: 35 | 100.00%WindowsTerminal.exe!NonClientIslandWindow::GetTotalNonClientExclusiveSize L: 45 | 100.00%WindowsTerminal.exe!IslandWindow::MessageHandler L: 55 | 100.00%WindowsTerminal.exe!NonClientIslandWindow::MessageHandler L: 65 | 100.00%WindowsTerminal.exe!BaseWindow_IslandWindow_::WndProc ``` ``` L: 426 | 100.00%WindowsTerminal.exe!NonClientIslandWindow::OnSize L: 526 | 100.00%WindowsTerminal.exe!IslandWindow::OnResize L: 626 | 100.00%WindowsTerminal.exe!IslandWindow::MessageHandler L: 726 | 100.00%WindowsTerminal.exe!NonClientIslandWindow::MessageHandler L: 826 | 100.00%WindowsTerminal.exe!BaseWindow_IslandWindow_::WndProc ``` Loading dumps.
claunia added the In-PRNeeds-Tag-FixProduct-TerminalSeverity-Crash labels 2026-01-31 06:59:42 +00:00
claunia commented 2026-01-31 06:59:44 +00:00
Author
Owner
Copy Link

@zadjii-msft commented on GitHub (May 26, 2023):

hey @lhecker member when you were like

"We should SetWindowLongPtr so that we don't get callbacks on our window message loops anymore"

yea we should probably do that. I'm not sure how to repro this, so this is gonna be one of them hypothetical fixes.

But all of these stacks are north of WindowThread::RundownForExit, where we're pumping remaining callbacks.

  • We close our XAML source,
  • we don't null out the IslandWindow,
  • we pump the remaining messages,
    • while we're pumping, we get the _WindowInitializedHandler coroutine scheduled
    • that calls ShowWindow,
    • that immediately calls back into NCIW / IslandWindow, based off the BaseWindow's WndProc.
      • In the case of GetTotalNonClientExclusiveSize, we try to ask for the _titlebar's height but OOPS NO XAML
      • In the case of OnSize, we call to SetWindowPos(_interopWindowHandle, but OOPS XAML ALREADY CloseWindow'd that HWND

So yea, I think we can definitely just sidestep this entire class of issues by doing a SetWindowLongPtr in IslandWindow::Close

@zadjii-msft commented on GitHub (May 26, 2023): hey @lhecker member when you were like "We should `SetWindowLongPtr` so that we don't get callbacks on our window message loops anymore" yea we should probably do that. I'm not sure how to repro this, so this is gonna be one of them _hypothetical fixes_. But all of these stacks are north of `WindowThread::RundownForExit`, where we're pumping remaining callbacks. * We close our XAML source, * we _don't_ null out the `IslandWindow`, * we pump the remaining messages, * while we're pumping, we get the `_WindowInitializedHandler` coroutine scheduled * that calls `ShowWindow`, * that immediately calls back into NCIW / IslandWindow, based off the `BaseWindow`'s `WndProc`. * In the case of `GetTotalNonClientExclusiveSize`, we try to ask for the `_titlebar`'s height but OOPS NO XAML * In the case of `OnSize`, we call to `SetWindowPos(_interopWindowHandle`, but OOPS XAML ALREADY `CloseWindow`'d that HWND So yea, I think we can definitely just sidestep this entire class of issues by doing a `SetWindowLongPtr` in `IslandWindow::Close`
claunia commented 2026-01-31 06:59:44 +00:00
Author
Owner
Copy Link

@lhecker commented on GitHub (May 26, 2023):

Oh, wow I had already forgotten about the SetWindowLongPtr approach.

@lhecker commented on GitHub (May 26, 2023): Oh, wow I had already forgotten about the `SetWindowLongPtr` approach.
claunia commented 2026-01-31 06:59:44 +00:00
Author
Owner
Copy Link

@zadjii-msft commented on GitHub (Jun 16, 2023):

Don't see any more of these on 1.18.1462. Huzzah?

@zadjii-msft commented on GitHub (Jun 16, 2023): Don't see any more of these on 1.18.1462. Huzzah?
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 In-PR Needs-Tag-Fix Product-Terminal 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#19989
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?