ags app launcher crashes hyprland with commit e989a0bcffac81092ed2a7e371f5225c113f689d #7127
Comments
|
fix 3 issues gtk will come back to haunt you can you post a debug stacktrace |
|
also @MightyPlaza I blame you, your fault |
|
is there a crash report? |
|
asked for one, we are waiting |
|
I looked for it but i did not find any that matches the correct date. |
|
debug stacktrace it is then |
|
i also tried following the steps for obtaining debug stack trace but coredumpctl says there aren't any |
|
disable crash reporter ( |
|
following the wiki + no crash reporter it still says no match found |
|
what should i do? |
|
can you send the full logs running ags only after hyprland has started? |
|
This is the log by running hyprland debug build, using wezterm to run |
|
does commenting your window rules change anything? |
|
it still crashes, i tried leaving only my inputs and keybinds related configuration modified while i used the default values for the rest but it also crashed |
|
idk if this helps |
|
does running ags with ayulr's ags config and then using the ags app launcher to try run an app also causes the crash for you guys? im not having any luck with logs or stack traces :( |
diff --git a/src/Compositor.cpp b/src/Compositor.cpp
index b0af0e9b..f2f58814 100644
--- a/src/Compositor.cpp
+++ b/src/Compositor.cpp
@@ -2255,16 +2255,25 @@ void CCompositor::updateFullscreenFadeOnWorkspace(PHLWORKSPACE pWorkspace) {
}
void CCompositor::changeWindowFullscreenModeInternal(const PHLWINDOW PWINDOW, const eFullscreenMode MODE, const bool ON) {
+ if (!PWINDOW)
+ return;
+
setWindowFullscreenInternal(
PWINDOW, (eFullscreenMode)(ON ? (uint8_t)PWINDOW->m_sFullscreenState.internal | (uint8_t)MODE : ((uint8_t)PWINDOW->m_sFullscreenState.internal & (uint8_t)~MODE)));
}
void CCompositor::changeWindowFullscreenModeClient(const PHLWINDOW PWINDOW, const eFullscreenMode MODE, const bool ON) {
+ if (!PWINDOW)
+ return;
+
setWindowFullscreenClient(PWINDOW,
(eFullscreenMode)(ON ? (uint8_t)PWINDOW->m_sFullscreenState.client | (uint8_t)MODE : ((uint8_t)PWINDOW->m_sFullscreenState.client & (uint8_t)~MODE)));
}
void CCompositor::setWindowFullscreenInternal(const PHLWINDOW PWINDOW, const eFullscreenMode MODE) {
+ if (!PWINDOW)
+ return;
+
if (PWINDOW->m_sWindowData.syncFullscreen.valueOrDefault())
setWindowFullscreenState(PWINDOW, sFullscreenState{.internal = MODE, .client = MODE});
else
@@ -2272,6 +2281,9 @@ void CCompositor::setWindowFullscreenInternal(const PHLWINDOW PWINDOW, const eFu
}
void CCompositor::setWindowFullscreenClient(const PHLWINDOW PWINDOW, const eFullscreenMode MODE) {
+ if (!PWINDOW)
+ return;
+
if (PWINDOW->m_sWindowData.syncFullscreen.valueOrDefault())
setWindowFullscreenState(PWINDOW, sFullscreenState{.internal = MODE, .client = MODE});
elseany difference? |
|
i tried it and i'm still crashing with the patch |
|
can you send the logs of doing the same thing on a version where it doesn't crash? |
|
this is the log where i open wezterm start logging with hyprctl, run ags and open librewolf: the problem is also present on my laptop, i tried getting the stack trace but coredumpctl says something .journal is truncated, googling suggests to increase size limit which i did from 32G to 64G but when i tried it again it said the same thing |
|
Not sure if related, but Hyprland is crashing for me (on commit Happy to make a separate issue and provide further info or try patches in a few hours, but here's the log for the time being in case it's relevant. Crash ( No Crash ( |
|
For me it crashes when I open up an Electron app (VS Code, Discord). I am on Edit: Nautilus also crashes Here is the last part of the log. Dunno how to upload a file without Drag-and-Drop :D |
|
so you wanna tell me that all of you can't get a debug stacktrace? fwiw I am running latest rn and not crashing at all lole |
|
I am also experiencing crashes. |
|
thanks, reprod |
|
fixed that one, is this done or do we have more than 1 bug here? |
|
i can confirm that in my case the culprit is |
|
I have |
Surprisingly yeah - no output from Probably not relevant, but when building (regular or debug), I get: And when doing Build does succeed with Can confirm that just odd there wasn't a way to quickly identify this. |
I can confirm that i am also experiencing where armcord which is an electron based app crash when i open it and it is most probably its own bug. |
|
debug stacktrace? I am not |
|
for me the electron based app (armcord) does not crash hyprland, it does its loading animation thingy and then disappears. is it the same for you @Visual-Dawg? |
|
in my case the other bug which makes armcord (an electron based app) to crash is caused by the commit 37e1411 according to the bisect armcord output: |
|
I use Vesktop, but VS Code and Vesktop instantly bring Hyprland to crash. I struggle a bit with a debug trace as my TTY currently gets spammed with warning messages, making it rather daunting to work with it. I'll try in the evening though |
|
@BluewyDiamond post this is solved, closing |
Regression?
Yes
System Info and Version
System/Version info
Description
Bisecting shows this commit e989a0b is the culpurit. In tty it says: terminated by signal SIGSEGV (Address boundary error).
I'm using ayulr's ags config with some minor tweaks.
How to reproduce
Crash reports, logs, images, videos
No response
The text was updated successfully, but these errors were encountered: