notepad-plus-plus
Strange crash - possible cause AMD Software Adrenalin Edition
I have a strange crash that is triggered when you try to move a docked dialog. It is triggered when the mouse button is released. The really strange thing is that I downloaded different Npp versions, up to version 8.0.0, and can reproduce this with different plugins that provide a docked dialog. But I'm relatively sure that I used this function two or three weeks ago and there was no crash.
As I said, I tested several versions (about 10) and was able to reproduce this behavior in all of them. I also tested various plugins: my own, PythonScript, NppExec, and the NppDemoPlugin. They all behaved the same way. It also doesn't matter whether I start Npp via Explorer, a CMD shell, or, as in the video, via Q-Dir.
At the moment, I can only imagine that this could have been caused by a Windows update, but that would be a bummer.
Here is the debug information
Notepad++ v8.8.7 (64-bit)
Build time: Oct 19 2025 - 16:19:15
Scintilla/Lexilla included: 5.5.7/5.4.5
Boost Regex included: 1_85
Path: C:\temp\deleteme\npp.8.8.7.portable.x64\notepad++.exe
Command Line:
Admin mode: OFF
Local Conf mode: ON
Cloud Config: OFF
Periodic Backup: ON
Placeholders: OFF
Scintilla Rendering Mode: SC_TECHNOLOGY_DIRECTWRITE (1)
Multi-instance Mode: monoInst
asNotepad: OFF
File Status Auto-Detection: cdEnabledNew (for current file/tab only)
Dark Mode: OFF
Display Info:
primary monitor: 1920x1080, scaling 100%
visible monitors count: 2
installed Display Class adapters:
0000: Description - AMD Radeon(TM) Graphics
0000: DriverVersion - 32.0.21033.2001
OS Name: Windows 11 Pro (64-bit)
OS Version: 24H2
OS Build: 26100.6899
Current ANSI codepage: 1252
Plugins:
mimeTools (3.1)
NppConverter (4.7)
NppExport (0.4)
NppPluginDemo (4.4)
I cannot reproduce the crash at all.
To be honest, I would be surprised if it could be reproduced so easily. By the way: For me, the easiest way to reproduce it is to simply double-click on the docked dialog to detach it so that it floats.
When I run it under the debugger, it reports that
int DockingCont::searchPosInTab(tTbData* pTbData) fails,
but ... that doesn't make sense ... there must be another reason.
As I said, I also test this functionality when I make changes to my plugins, and it was still working a few weeks ago. And even if a problem has crept in now, why should it also be present in version 8.0.0? And why with different dialogs?
Here the config.
Cannot reproduce either.
Could you show me also pic of your callstack? (just at the same DockingCont::searchPosInTab point as in the pic above)
Hello, @donho, @ekopalypse and All,
First, I must say that I migrated to Windows 11 version 25H2 on October, 25. I also noticed some weird behaviors with my portable Notepad++ v8.8.6 release. However, my portable N++ v8.8.7 is working correctly, so far !
I suppose, @ekopalypse and @donho, that it has some relation with the rendering mode :-((
On an USB key, I had two folders 886_x64 and 887_x64 containing the associated N++ version
-
From a DOS prompt, I ran the command
D:\886_x64>notepad++ -nosession -noPlugin change.log -
Once in N++, I opened the Find dialog and searched for the word
Notepadin all theChange.logcontents
When I dragged the Find Result panel to the center of the screen it did not crash but the screen flickered between the change.log text and a black screen. and when I put back the Find results panel to its usual location, the screen flickered again between the change.log text and a black screen :-)) See below :
https://github.com/user-attachments/assets/3f2d4e67-b802-4be3-b2fe-fd262ebc08b8
The debug info :
Notepad++ v8.8.6 (64-bit) Build time: Oct 1 2025 - 01:23:06 Scintilla/Lexilla included: 5.5.7/5.4.5 Boost Regex included: 1_85 Path: D:\886_x64\notepad++.exe Command Line: -nosession -noPlugin change.log Admin mode: OFF Local Conf mode: ON Cloud Config: OFF Periodic Backup: OFF Placeholders: OFF Scintilla Rendering Mode: SC_TECHNOLOGY_DIRECTWRITE (1) Multi-instance Mode: monoInst asNotepad: OFF File Status Auto-Detection: cdEnabledNew (for current file/tab only) Dark Mode: OFF Display Info: primary monitor: 1920x1080, scaling 125% visible monitors count: 1 installed Display Class adapters: 0001: Description - Intel(R) Iris(R) Xe Graphics 0001: DriverVersion - 32.0.101.6913 OS Name: Windows 11 Pro (64-bit) OS Version: 25H2 OS Build: 26200.7019 Current ANSI codepage: 1252 Plugins: none
Note that this behavior occurred ONLY IF the rendering mode is set to Direct Write (default) or Direct Write (Direct X11), so for cases 1 and 4 in the Debug Info contents. For all the other rendering modes ( 0, 2 and 3 ), everything was fine ?
Now, running, from a DOS prompt, the command D:\887_x64>notepad++ -nosession -noPlugin change.log, everything was fine, in any rendering mode !
Best Regards,
guy038
@Ekopalypse I couldn't get the same callstack at all when trying to imitate your crash STR. It means that there must be an external influence (a SW or HW condition change) redirecting the relevant panel undocking code-flow somehow. There are many relevant branching conditions depending on the current graphics (e.g. current dragging mouse position calc on monitor or RECT sizes of wnds, etc.). I noticed this in your DebugInfo:
visible monitors count: 2
Could you try to disconnect one of the monitors and repeat your crash STR in the single mon-mode (I'd try both monitors, one by one)?
I tested the single-monitor solution, but the result was the same. Then I uninstalled the AMD Graphics Suite software, as I saw that it had been updated recently. After restarting, the problem persisted. I then uninstalled the latest Windows updates, managing to uninstall two of them but not the third. After restarting, I ran the test again, but without success. I also tested GDI instead of DirectX, but unfortunately without success. The hardware has not changed since I started using Windows 11 (~6 months). I hope this is a local problem on my end and not one generated by a Windows update, because that would really suck.
Hi, @ekopalypse,
To be rigorously exact, the first time I tried to do your STR with my portable 8.8.6 release, my Windows 11 laptop did crash and I had the usual message that the problem was transmitted to Microsoft !
But, surprisingly, trying again your STR, a couple of minutes after, it did not crash and just showed the flickered screen mentioned in my previous post
Thus, it looks like my system has understood the problem and proposed, somehow, an improvement !!??
BR
guy038
Hello @guy038 Did the operating system itself crash or just Notepad++? Since I've been testing it, I've never seen a dialog box indicating that this program crash is being reported to MS. The more I think about it, the more I am convinced that it happened after the latest cumulative update for Windows 11, the 2025-10. But this update updates the boot certificates, so I am hesitant to uninstall it, even though it says that the certificates won't expire until 2026... hmm.
Hi, @ekopalypse,
No, the system did not crash, just Notepad++. Cancelling N++ just triggers the transmission to Microsoft.
BR
guy038
Hello, @ekopalypse,
Here is a snapshot of my last updates on my Windows 11 ( ex Windows 10 ! ) laptop :
BR
guy038
I'm still on Windows 11, version 24H2... I'll probably give it a try, even though I don't really like what it has to “offer.”
@Ekopalypse I was finally able to obtain the same callstack (and thus the same code-flow) as is in your pic.
From your 1st post video and your report:
move a docked dialog. It is triggered when the mouse button is released.
is obvious that you are dragging the panel and the crash occurs when you release it.
But your callstack is like you doubleclicked the panel caption (to undock it)!
Did you changed somehow settings of your mouse? Could you replace your current one and test with another mouse?
Did you changed somehow settings of your mouse?
No, and my mouse is really simple—it doesn't need any software because it can't do anything except click and scroll :-)
Could you replace your current one and test with another mouse?
There must be one lying around somewhere. If not, I'll steal one from my colleague. :-D But that would take a while until the first one goes home.
I performed the update to 25H2, but that didn't solve the problem either. :-(
The Logi M185 (wireless) was replaced with a G403 (USB cable), but this did not solve the problem.
There are many amdxx64 threads that look suspicious, but honestly, I have no experience with whether this could indicate a problem.
But as you can see, the crash place substantially changed. Now there is the N++ Gripper visible as it should in such a dragging case.
There are many amdxx64 threads that look suspicious
That's ok, it's a OS thing, not a N++. This is how it nowadays in Windows works. The only real N++ threads are the main one (TID 0x1ec8) and the backup one (TID 0x3968).
@Ekopalypse Ok, let's narrow down the possible culprits. If you repeat this N++ issue in a fresh Win11 VM on your comp, you will be able to decide if it's a HW or a SW problem. If it works ok in the VM, it's in your current Windows settings/SW. If not, it's most probably a HW problem on your comp.
I haven't set up any VM environments yet – but yes, that could lead to some conclusions. It's hard to believe that this is a hardware problem, as I would assume that other applications would also be affected and that other actions would also cause crashes, not just this one specific one. I will conduct further tests to see if the problem can be isolated in any way. I am currently learning how to use this new “time travel debugging” feature, so perhaps that will shed some light on the matter.
Nevertheless, thank you all for your help and the time you have spent on this strange problem.
... and imagine, I set up the sandbox as a quick test and it works there.
... and imagine, I set up the sandbox as a quick test and it works there.
Good, so now you know that's in your current SW env. Do you use system restore points? If so, and you have an older restore point from the past where you are sure everything was working properly, you can go back in time (old Windows Control Panel > Recovery > Open System Restore > Choose a different restore point...).
Let us know here if you find the culprit, I'm curious :-)
Let us know here if you find the culprit
I definitely will
I'm curious
me too :-D
Hello, @ekopalypse, @xomx and All,
Some additionnal points about this bug :
-
First, as I usually type with a
wirelessMicrosoft keyboard and navigate with awirelessMicrosoft mouse, I did a try with a NECUSBmouse but no change at all ! -
Secondly, I did install the
NppPluginDemoplugin of Don and the behavior, noticed in my first post, is exactly the same with this plugin ! See below :
https://github.com/user-attachments/assets/52330e8b-08a6-4456-8c7f-d895f2c9d792
Thus, in summary :
-
With a portable
N++ v8.8.6 x64installation, it's fine with the0, 2 and 3rendering mode and the flickering behavior occurred with theDirectWrite (default)andDirectWrite (DirectX 11)rendering modes, so for cases1 or 4. -
With a portable
N++ v8.8.7 x64installation, everything is fine, whatever the rendering mode used. -
I could reproduce these behaviors using an other mouse device.
Best Regards
guy038
Hi, All,
Last news : unlike I pretended in my previous post, today, I've got exactly the same bug with my portable N++ v8.8.7 install, either :-((. Don't know why ?
BR
guy038
@guy038
I've got exactly the same bug with my portable
N++ v8.8.7install
I'm not surprised, there is no significant difference in between the v8.8.7 & v8.8.6 (regarding the graphics of the text editing window area).
installed Display Class adapters: 0001: Description - Intel(R) Iris(R) Xe Graphics 0001: DriverVersion - 32.0.101.6913
I'd try to update your Intel GPU driver. If your computer manufacturer doesn't offer one, here you will find the generic Intel one: https://www.intel.com/content/www/us/en/search.html#sort=%40lastmodifieddt%20descending&f:@tabfilter=[Downloads]&f:@stm_10385_en=[Graphics,Processor%20Graphics,Intel%C2%AE%20Iris%C2%AE%20Xe%20Graphics%20Family]
Hello, @xomx
First, thank you very much for the link for the Intel driver !
Now, initially, I preferred to go to the HP support site and giving my serial number, it did find my laptop !
Then, I chose the option Let us detect the drivers you need for this HP computer
Here is the resulting list of proposed drivers, translated in English, of course :
Welcome to software and drivers for 'HP ProBook 450 G8 Notebook PC' (2W8T5EA)
Selected operating system : Windows 11 version 25H2 (64 bits)
Detection complete. Here are the software and drivers available for your system:
• Display Link graphic driver Optional 11.7.5493.0 Rev.A 8.9 MB July 9, 2025 Download
Type: Graphics Card Driver
Version: 11.7.5493.0 Rev.A
Operating Systems: Windows 11 version 25H2 (64-bit)
Release Date: July 9, 2025
File name: sp160418.exe (8.8 MB)
Definition: This package contains the Display Link graphics driver for compatible computers running a supported operating system
Fixes and improvements: Provides the updated driver
• HP Application Activation Software Driver Optional 1.80.4268.0 Rev.A 113.6 MB July 16, 2025 Download
• HP BIOS and System Firmware (T70/T85/T87/T88/T89/T99) Optional 01.22.00 Rev.A 109.4 MB Sept. 22, 2025 Download
• HP Connection Optimizer Optional 2.0.20.0 Rev.A 4.1 MB April 25, 2023 Download
• HP Thunderbolt Dock G4 - Firmware Optional 1.5.27.0 Rev.C 12.2 MB May 13, 2025 Download
• HP USB-C Dock G5 - Firmware Optional 1.0.25.0 Rev.C 13.7 MB July 18, 2025 Download
• HP USB-C Dock G5 - Audio Driver Optional 6.4.0.2412/393 Rev.A 19.1 MB April 15, 2025 Download
• HP USB-C/A Dock G2 - Firmware Optional 1.1.24.0 Rev.B 13.5 MB July 18, 2025 Download
• HP XMM7360 WWAN Driver Optional 2.10.4.124 Rev.A 51.2 MB Aug. 21, 2023 Download
• Intel Bluetooth Driver Optional 23.160.0.9 Rev.A 10.2 MB Sept. 1, 2025 Download
• Intel Dynamic Tuning Optional 8.7.10802.26924 Rev.A 4.4 MB June 3, 2024 Download
• Intel HID Event Filter Driver Optional 2.2.2.9 Rev.A 0.8 MB Mar 26, 2024 Download
• Intel Management Engine Driver Optional 2514.7.16.0 Rev.A 147.6 MB June 25, 2025 Download
• Intel Serial IO Driver Optional 30.100.2129.8 Rev.A 1.1 MB Oct. 14, 2021 Download
• Intel WLAN Driver Optional 23.160.0.4 Rev.A 14.3 MB Sept . 2, 2025 Download
• NVIDIA Video Driver and Control Panel Optional 32.0.15.7344 Rev.W 1153.9 MB July 24, 2025 Download
• Intel video driver and control panel Optional 32.0.101.6913 Rev.W 902.7 MB July 31, 2025 Download
Type: Graphics Card Driver
Version: 32.0.101.6913 Rev.W
Operating Systems: Windows 11 version 25H2 (64-bit)
Release Date: July 31, 2025
File name: sp161575.exe (902.6 MB)
Definition: This package contains Intel video drivers, control panel, and display audio driver for supported laptop models running a compatible operating system
Fixes and improvements: Provides the updated Intel UHD/HD graphics driver
• Realtek Ethernet Controller Driver Optional 1166.15.717.2023 Rev.A 1.1 MB Feb. 26, 2024 Download
• Realtek HD Audio Driver Optional 15.9844.9947.95 Rev.Q 366.4 MB July 11, 2025 Download
• Realtek RTL8xxx Wireless LAN Driver Optional 1.0.0.267 Rev.S 17.1 MB June 20, 2025 Download
• Realtek USB Audio Driver for HP G2 Docking Station Optional 6.4.0.393 Rev.A 18.2 MB July 10, 2024 Download
• System Default Settings Optional 1.5.9.1 Rev.0 4.5 MB June 10, 2025 Download
• Intel Chipset Installation Utility Optional 10.1.19627.8423 Rev.A 0.9 MB Mar 7, 2024 Download
I thought that the only interesting software were Display Link graphic driver and Intel video driver and control panel
I also noticed that, in the history of the installed drivers, the two references :
-
Intel Corporation - Display - 32.0.101.6913installed on Oct. 25, 2025 -
Intel Corporation - Display - 32.0.101.6913installed on Sept. 27, 2025
Thus, I decided to run these two installation files :
-
sp160418.exe - 11.7.5493.0 Rev.Aof July 9, 2025 -
sp161575.exe - 32.0.101.6913 Rev.Wof July 31, 2025
Unfortunately, even after a laptop restart, no changes and the bug remains !
Now, from your link, it seems that this Intel version is more recent than all my present versions. The version number 32.0.101.7076 is, indeed, superior to 32.0.101.6913 :
Intel® 11th – 14th Gen Processor Graphics - Windows Date: 09/19/2025 Version: 32.0.101.7076
https://www.intel.com/content/www/us/en/download/864990/intel-11th-14th-gen-processor-graphics-windows.html
I downloaded the gfx_win_101.7076.exe file but did not run it, so far. I suppose it's quite safe, isn't ?
Best Regards,
guy038
@guy038
I downloaded the
gfx_win_101.7076.exefile but did not run it, so far. I suppose it's quite safe, isn't ?
It depends on what you mean by the "safe". I used such generic installers from the Intel many times for my Intel-GPU comps without a problem but every driver installation is a risk, you can end up with a non-functional PC, you are on your own with this. You can minimize the risk by using the System Restore feature (always creating a Restore Point before such an op).
Hi, @xomx,
I followed your instructions and I did create a restore point that I named AVANT exécution de 'gfx_win_101.7076.exe' ( Intel update ) !
Then, I installed the gfx_win_101.7076.exe program, around 14h15 and everything went smoothly. It even did a new update of the graphic pilot from version 32.0.101.7076 to version 32.0.101.7080
However, after checking with my portable 8.8.7 N++, the display bug still persists ! Again, on my laptop, the rendering modes 0,2 and 3 work correctly and the rendering modes 1 and 4 trigger the flickering :-((
BTW, here is a snapshot of the last part of my installed applications list, sorted by installation date :
Thus, do you think that I can keep this Intel update or would it be better to use my previous restore point ? Note that I did not notice any deterioration of my current display, regarding colors and quality, using other software !
BR
guy038
P.S. : I don't know if all the Microsoft installations of Edge Desktop Runtime, Visual C++ and ASP Net are consecutive to the installation of the new Intel graphic driver but they occur on the same day than the installation. However, I did not found any new reference in Windows Update either !
@guy038
do you think that I can keep this Intel update or would it be better to use my previous restore point ?
I'd keep it. If you read the Detailed Description from your link
Intel® 11th – 14th Gen Processor Graphics - WindowsDate:09/19/2025Version:32.0.101.7076https://www.intel.com/content/www/us/en/download/864990/intel-11th-14th-gen-processor-graphics-windows.html
it's clear that from now on there will be a very small chance that your OEM (HP) will ever update its drivers:
Note: Intel has moved 11th - 14th Gen Intel® Processor Graphics and related Intel Atom®, Pentium®, and Celeron® processor
graphics to a legacy software support model. For more information on this support update and additional changes to the driver
package, see [here](https://www.intel.com/content/www/us/en/support/articles/000101986/graphics.html).
And the possible problems, with such a generic driver, is described right in the next paragraph:
The generic Intel® driver provides users with the latest and greatest feature enhancements and bug fixes that computer
manufacturers (OEMs) might not have customized yet. OEM drivers are handpicked and include customized features and solutions
to platform-specific needs. Installing this Intel generic graphics driver will overwrite your handpicked OEM graphics driver.
Users can check for matching OEM driver versions at OEM websites. For more information on how the installation of this driver
may impact your OEM customizations, see this [article.](https://www.intel.com/content/www/us/en/support/articles/000096252/graphics.html)
Hello, @xomx and All,
So, as you said, I think that the sensible way is to keep this Intel update. Thank you, again, for helping me through this whole process !
BTW, I've found out a very good on-line screen tester, with many tests and many options for each test :
https://display-test.app/
After running all the tests, my screen does not seem to have any default or dead pixel ! Apart from the Dead Pixel test, I particularly like the following tests :
-
Color Gradient test
-
Color Distance test
-
Gamma test
-
Test Patterns (
80tests all different ! )
Regarding the Color gradient test, I can distinguish between two colors even if the three colors (RGB) differ, ONLY by two units each. For a one unit difference, it's not so obvious !
The site https://www.eizo.be/monitor-test/ is also quite valuable. Their Gamma test is very easy to realize : it said that my screen Gamma is 1,3, which seems a poor value !
Note that when you hit the Start button, it asks you to set the zoom factor of your browser and your desktop to 100 %. I do use the default zoom of 100 % for my browser, but, personally, I preferred to keep the recommended value for my screen, which is 125 % !
Best Regards
guy038
After I disabled all settings, the crash disappeared. What surprises me is that I had already completely uninstalled the software and restarted the PC, but the crash still happened... hmm... I'm leaving the issue open for now to see if it can actually be confirmed, but if you think it doesn't make sense, you can close it, of course.