starred/terminal
1
0
Fork 0
mirror of https://github.com/microsoft/terminal.git synced 2026-07-08 18:16:28 +00:00
Code Issues 19.0k Packages Projects Releases 20 Wiki Activity

Interactivity bugs post-#9820 #13633

New Issue
Closed
opened 2026-01-31 03:47:51 +00:00 by claunia · 8 comments
claunia commented 2026-01-31 03:47:51 +00:00
Owner
Copy Link

Originally created by @zadjii-msft on GitHub (Apr 26, 2021).

Originally assigned to: @zadjii-msft on GitHub.

This is a tracking thread for things I probably broke in #9820. I'd rather not file 100 bugs for these

Mysterious DX crash

I experienced a crash when I killed DWM, and it took particularly long to recover

crash was on:

            RETURN_IF_FAILED(_dxgiFactory2->CreateSwapChainForComposition(_d3dDevice.Get(),
                                                                          &_swapChainDesc,
                                                                          nullptr,
                                                                          &_dxgiSwapChain));

in

0:000> k
 # Child-SP          RetAddr               Call Site
00 00000043`cef8ec80 00007ff9`3a4ac5a1     Microsoft_Terminal_Control!Microsoft::Console::Render::DxEngine::_CreateDeviceResources+0x47b [E:\BA\242\s\src\renderer\dx\DxRenderer.cpp @ 630] 
01 (Inline Function) --------`--------     Microsoft_Terminal_Control!Microsoft::Console::Render::DxEngine::GetSwapChain+0x14 [E:\BA\242\s\src\renderer\dx\DxRenderer.cpp @ 1010] 
02 00000043`cef8ed50 00007ff9`3a55667d     Microsoft_Terminal_Control!winrt::Microsoft::Terminal::Control::implementation::ControlCore::GetSwapChain+0x21 [E:\BA\242\s\src\cascadia\TerminalControl\ControlCore.cpp @ 1183] 
03 00000043`cef8ed80 00007ff9`3a4e1f71     Microsoft_Terminal_Control!winrt::Microsoft::Terminal::Control::implementation::TermControl::RenderEngineSwapChainChanged$_ResumeCoro$1+0x2bd [E:\BA\242\s\src\cascadia\TerminalControl\TermControl.cpp @ 582] 
04 (Inline Function) --------`--------     Microsoft_Terminal_Control!std::experimental::coroutine_handle<void>::resume+0x9 [C:\Program Files (x86)\Microsoft Visual Studio\2019\Enterprise\VC\Tools\MSVC\14.28.29910\include\experimental\coroutine @ 107] 
05 (Inline Function) --------`--------     Microsoft_Terminal_Control!std::experimental::coroutine_handle<void>::operator()+0x9 [C:\Program Files (x86)\Microsoft Visual Studio\2019\Enterprise\VC\Tools\MSVC\14.28.29910\include\experimental\coroutine @ 99] 
06 00000043`cef8ee50 00007ff9`5e2623a6     Microsoft_Terminal_Control!winrt::impl::resume_apartment_callback+0x11 [E:\BA\242\s\src\cascadia\TerminalControl\Generated Files\winrt\windows.ui.core.h @ 1088] 
07 00000043`cef8ee80 00007ff9`5e2af93f     Windows_UI!<lambda_59517943c03487243f9bea31c6c1a784>::operator()+0x84 [onecoreuap\windows\advcore\winrt\onecoreiwindow\corewindow\common\dispatcher.cpp @ 908] 
08 00000043`cef8eeb0 00007ff9`7de6a440     Windows_UI!Microsoft::WRL::Details::DelegateArgTraits<long (__cdecl Windows::System::IDispatcherQueueHandler::*)(void)>::DelegateInvokeHelper<Microsoft::WRL::Implements<Microsoft::WRL::RuntimeClassFlags<2>,Windows::System::IDispatcherQueueHandler,Microsoft::WRL::FtmBase>,<lambda_59517943c03487243f9bea31c6c1a784>,-1>::Invoke+0xf [onecore\external\sdk\inc\wrl\event.h @ 354] 
09 00000043`cef8eee0 00007ff9`7de4229c     CoreMessaging!Windows::System::DispatcherQueue::DeferInvokeCallback+0x20 [mincore\coreui\dev\dispatcherqueue\wrtdispatcherqueue.cpp @ 903] 
0a (Inline Function) --------`--------     CoreMessaging!Microsoft::CoreUI::ActionCallback::ImportAdapter$::__l2::<lambda_a81ff790741c2a62f2197c2561f5fe49>::operator()+0x1f [mincore\CoreUI\Dev\System\Api\IExportMessageSession.cs @ 22]

the _dxgiFactory2 was null. That's unexpected. It shouldn't be null.

Originally created by @zadjii-msft on GitHub (Apr 26, 2021). Originally assigned to: @zadjii-msft on GitHub. This is a tracking thread for things I probably broke in #9820. I'd rather not file 100 bugs for these * [x] (**A**) Mousedown, select, SCROLL does a weird thing with endpoints that doesn't happen in stable - fixed in 4f4df01 - fixed in #9980 * [x] (**B**) Trackpad scrolling _down_ with _small_ increments seems buggy. But scrolling with a big increment goes fast? And scrolling with the mouse is fine. And trackpad scrolling in mouse mode is fine. - fixed in #9980 * [x] (**C**) Looks like there's a selection issue when you click and drag too quickly. Seems like the start endpoint doesn't always get set. - fixed in 9d33e89 - fixed in #9980 * [x] (**D**) Mysterious DX crash - _this never reprod, so we're assuming it's unrelated_ <details> <summary>Mysterious DX crash</summary> I experienced a crash when I killed DWM, and it took particularly long to recover crash was on: ```c++ RETURN_IF_FAILED(_dxgiFactory2->CreateSwapChainForComposition(_d3dDevice.Get(), &_swapChainDesc, nullptr, &_dxgiSwapChain)); ``` in ``` 0:000> k # Child-SP RetAddr Call Site 00 00000043`cef8ec80 00007ff9`3a4ac5a1 Microsoft_Terminal_Control!Microsoft::Console::Render::DxEngine::_CreateDeviceResources+0x47b [E:\BA\242\s\src\renderer\dx\DxRenderer.cpp @ 630] 01 (Inline Function) --------`-------- Microsoft_Terminal_Control!Microsoft::Console::Render::DxEngine::GetSwapChain+0x14 [E:\BA\242\s\src\renderer\dx\DxRenderer.cpp @ 1010] 02 00000043`cef8ed50 00007ff9`3a55667d Microsoft_Terminal_Control!winrt::Microsoft::Terminal::Control::implementation::ControlCore::GetSwapChain+0x21 [E:\BA\242\s\src\cascadia\TerminalControl\ControlCore.cpp @ 1183] 03 00000043`cef8ed80 00007ff9`3a4e1f71 Microsoft_Terminal_Control!winrt::Microsoft::Terminal::Control::implementation::TermControl::RenderEngineSwapChainChanged$_ResumeCoro$1+0x2bd [E:\BA\242\s\src\cascadia\TerminalControl\TermControl.cpp @ 582] 04 (Inline Function) --------`-------- Microsoft_Terminal_Control!std::experimental::coroutine_handle<void>::resume+0x9 [C:\Program Files (x86)\Microsoft Visual Studio\2019\Enterprise\VC\Tools\MSVC\14.28.29910\include\experimental\coroutine @ 107] 05 (Inline Function) --------`-------- Microsoft_Terminal_Control!std::experimental::coroutine_handle<void>::operator()+0x9 [C:\Program Files (x86)\Microsoft Visual Studio\2019\Enterprise\VC\Tools\MSVC\14.28.29910\include\experimental\coroutine @ 99] 06 00000043`cef8ee50 00007ff9`5e2623a6 Microsoft_Terminal_Control!winrt::impl::resume_apartment_callback+0x11 [E:\BA\242\s\src\cascadia\TerminalControl\Generated Files\winrt\windows.ui.core.h @ 1088] 07 00000043`cef8ee80 00007ff9`5e2af93f Windows_UI!<lambda_59517943c03487243f9bea31c6c1a784>::operator()+0x84 [onecoreuap\windows\advcore\winrt\onecoreiwindow\corewindow\common\dispatcher.cpp @ 908] 08 00000043`cef8eeb0 00007ff9`7de6a440 Windows_UI!Microsoft::WRL::Details::DelegateArgTraits<long (__cdecl Windows::System::IDispatcherQueueHandler::*)(void)>::DelegateInvokeHelper<Microsoft::WRL::Implements<Microsoft::WRL::RuntimeClassFlags<2>,Windows::System::IDispatcherQueueHandler,Microsoft::WRL::FtmBase>,<lambda_59517943c03487243f9bea31c6c1a784>,-1>::Invoke+0xf [onecore\external\sdk\inc\wrl\event.h @ 354] 09 00000043`cef8eee0 00007ff9`7de4229c CoreMessaging!Windows::System::DispatcherQueue::DeferInvokeCallback+0x20 [mincore\coreui\dev\dispatcherqueue\wrtdispatcherqueue.cpp @ 903] 0a (Inline Function) --------`-------- CoreMessaging!Microsoft::CoreUI::ActionCallback::ImportAdapter$::__l2::<lambda_a81ff790741c2a62f2197c2561f5fe49>::operator()+0x1f [mincore\CoreUI\Dev\System\Api\IExportMessageSession.cs @ 22] ``` the `_dxgiFactory2` was null. That's unexpected. It shouldn't be null. </details> * [x] (**E**) It doesn't seem like we're calling this when we create the automation peer -- we are not guaranteed to always call UpdateSettings after creating a peer, so the first peer may have no padding. See https://github.com/microsoft/terminal/pull/10051#discussion_r672443301. (fixed by #10874)
claunia commented 2026-01-31 03:47:54 +00:00
Author
Owner
Copy Link

@DHowett commented on GitHub (Apr 28, 2021):

Looks like there's a selection issue when you click and drag too quickly.

(Sorry for the terse body; I have to post this using powershell because of a new MSIT rule about personal devices :|)

@DHowett commented on GitHub (Apr 28, 2021): Looks like there's a selection issue when you click and drag too quickly. (Sorry for the terse body; I have to post this using powershell because of a new MSIT rule about personal devices :|)
claunia commented 2026-01-31 03:47:54 +00:00
Author
Owner
Copy Link

@zadjii-msft commented on GitHub (Apr 28, 2021):

Looks like there's a selection issue when you click and drag too quickly.

There sure is, just hit like 4 times this morning. Added to the list ☺️

@zadjii-msft commented on GitHub (Apr 28, 2021): > Looks like there's a selection issue when you click and drag too quickly. There sure is, just hit like 4 times this morning. Added to the list ☺️
claunia commented 2026-01-31 03:47:54 +00:00
Author
Owner
Copy Link

@zadjii-msft commented on GitHub (Apr 28, 2021):

haha, found it

if (_singleClickTouchdownPos)
{
    // Figure out if the user's moved a quarter of a cell's smaller axis away from the clickdown point
    auto& touchdownPoint{ *_singleClickTouchdownPos };
    auto distance{ std::sqrtf(std::powf(cursorPosition.X - touchdownPoint.X, 2) + std::powf(cursorPosition.Y - touchdownPoint.Y, 2)) };
    const til::size fontSize{ _actualFont.GetSize() };

    const auto fontSizeInDips = fontSize.scale(til::math::rounding, 1.0f / _renderEngine->GetScaling());
    if (distance >= (std::min(fontSizeInDips.width(), fontSizeInDips.height()) / 4.f))
    {
        _terminal->SetSelectionAnchor(_GetTerminalPosition(touchdownPoint));
        // stop tracking the touchdown point
        _singleClickTouchdownPos = std::nullopt;
    }
}

if (_singleClickTouchdownPos)
{
    // Figure out if the user's moved a quarter of a cell's smaller axis away from the clickdown point
    auto& touchdownPoint{ *_singleClickTouchdownPos };
    float dx = ::base::saturated_cast<float>(pixelPosition.x() - touchdownPoint.x());
    float dy = ::base::saturated_cast<float>(pixelPosition.y() - touchdownPoint.y());
    auto distance{ std::sqrtf(std::powf(dx, 2) +
                              std::powf(dy, 2)) };

    const auto fontSizeInDips{ _core->FontSizeInDips() };
    if (distance >= (std::min(fontSizeInDips.width(), fontSizeInDips.height()) / 4.f))
    {
        _core->SetSelectionAnchor(terminalPosition);
        // stop tracking the touchdown point
        _singleClickTouchdownPos = std::nullopt;
    }
}

        _terminal->SetSelectionAnchor(_GetTerminalPosition(touchdownPoint));

vs

        _core->SetSelectionAnchor(terminalPosition);

We're now using the location of the drag event as the selection anchor, instead of the location that the user initially clicked. Oops. Now to write a test.

@zadjii-msft commented on GitHub (Apr 28, 2021): haha, found it <table> <tr> <td> ```c++ if (_singleClickTouchdownPos) { // Figure out if the user's moved a quarter of a cell's smaller axis away from the clickdown point auto& touchdownPoint{ *_singleClickTouchdownPos }; auto distance{ std::sqrtf(std::powf(cursorPosition.X - touchdownPoint.X, 2) + std::powf(cursorPosition.Y - touchdownPoint.Y, 2)) }; const til::size fontSize{ _actualFont.GetSize() }; const auto fontSizeInDips = fontSize.scale(til::math::rounding, 1.0f / _renderEngine->GetScaling()); if (distance >= (std::min(fontSizeInDips.width(), fontSizeInDips.height()) / 4.f)) { _terminal->SetSelectionAnchor(_GetTerminalPosition(touchdownPoint)); // stop tracking the touchdown point _singleClickTouchdownPos = std::nullopt; } } ``` </td> <td> ```c++ if (_singleClickTouchdownPos) { // Figure out if the user's moved a quarter of a cell's smaller axis away from the clickdown point auto& touchdownPoint{ *_singleClickTouchdownPos }; float dx = ::base::saturated_cast<float>(pixelPosition.x() - touchdownPoint.x()); float dy = ::base::saturated_cast<float>(pixelPosition.y() - touchdownPoint.y()); auto distance{ std::sqrtf(std::powf(dx, 2) + std::powf(dy, 2)) }; const auto fontSizeInDips{ _core->FontSizeInDips() }; if (distance >= (std::min(fontSizeInDips.width(), fontSizeInDips.height()) / 4.f)) { _core->SetSelectionAnchor(terminalPosition); // stop tracking the touchdown point _singleClickTouchdownPos = std::nullopt; } } ``` </td> </tr> </table> ```c++ _terminal->SetSelectionAnchor(_GetTerminalPosition(touchdownPoint)); ``` vs ```c++ _core->SetSelectionAnchor(terminalPosition); ``` We're now using the location of the drag event as the selection anchor, instead of the location that the user initially clicked. Oops. Now to write a test.
claunia commented 2026-01-31 03:47:54 +00:00
Author
Owner
Copy Link

@zadjii-msft commented on GitHub (Apr 28, 2021):

Oof. B is hard, I think. The touchpad sends very many small scroll deltas, less than one row at a time. The control scrollbar can store a double, so small deltas can accumulate. Originally, these would accumulate in the scrollbar, and we'd only read that out as an int in the scrollbar updater, which is throttled.

In the interactivity split, there's no place for us to store that double. We immediately narrow to an int for ControlInteractivity::_updateScrollbar.

Theoretically, we could store a double in Interactivity that's a fake scrollbar position, to accumulate to. Then, ControlInteractivity would need to listen to scroll events from the core, unsure how easy that is right now...

@zadjii-msft commented on GitHub (Apr 28, 2021): Oof. B is hard, I think. The touchpad sends very many small scroll deltas, less than one row at a time. The control scrollbar can store a `double`, so small deltas can accumulate. Originally, these would accumulate in the scrollbar, and we'd only read that out as an `int` in the scrollbar updater, which is throttled. In the interactivity split, there's no place for us to store that double. We immediately narrow to an `int` for `ControlInteractivity::_updateScrollbar`. Theoretically, we could store a `double` in `Interactivity` that's a fake scrollbar position, to accumulate to. Then, `ControlInteractivity` would need to listen to scroll events from the core, unsure how easy that is right now...
claunia commented 2026-01-31 03:47:54 +00:00
Author
Owner
Copy Link

@DHowett commented on GitHub (Apr 28, 2021):

Oh yes, that is frightening for sure.

This is a wackadoo idea, but... what if the scrollbar's size was WHEEL_DELTA * BufferHeight and we just ... let the deltas accumulate and only scroll every WHEEL_DELTA units

@DHowett commented on GitHub (Apr 28, 2021): Oh yes, that is frightening for sure. This is a wackadoo idea, but... what if the scrollbar's size was `WHEEL_DELTA * BufferHeight` and we just ... let the deltas accumulate and only scroll every WHEEL_DELTA units
claunia commented 2026-01-31 03:47:55 +00:00
Author
Owner
Copy Link

@DHowett commented on GitHub (Apr 28, 2021):

No no it would probably be absolutely wack for the UI to display that (or Accessibility would say "line 300 of 1991661")

@DHowett commented on GitHub (Apr 28, 2021): No no it would probably be absolutely wack for the UI to display that (or Accessibility would say "line 300 of 1991661")
claunia commented 2026-01-31 03:47:55 +00:00
Author
Owner
Copy Link

@DHowett commented on GitHub (Jul 19, 2021):

Point [E] may cause an accessibility regression in the 1.11 timeframe.

@DHowett commented on GitHub (Jul 19, 2021): Point [E] may cause an accessibility regression in the 1.11 timeframe.
claunia commented 2026-01-31 03:47:55 +00:00
Author
Owner
Copy Link

@ghost commented on GitHub (Aug 31, 2021):

:tada:This issue was addressed in #10874, which has now been successfully released as Windows Terminal Preview v1.11.2421.0.🎉

Handy links:

@ghost commented on GitHub (Aug 31, 2021): :tada:This issue was addressed in #10874, which has now been successfully released as `Windows Terminal Preview v1.11.2421.0`.:tada: Handy links: * [Release Notes](https://github.com/microsoft/terminal/releases/tag/v1.11.2421.0) * [Store Download](https://www.microsoft.com/store/apps/9n8g5rfz9xk3?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 Priority-1 Product-Terminal Resolution-Fix-Committed
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#13633
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?