ghostty crashes after resizing

[kjughx]

I have open ghostty with a vertical split, i.e. one left and one right split. In one split I run a python script which generates a new window, causing ghostty to resize, after closing the new window ghostty freezes for a second then crashes due to SEGV. It does not consistently happen so I don't have a reliable way to reproduce it unfortunately.

[mitchellh]

We'll need a reproduction to accept this.

[dariogriffo]

I can reproduce this on debian testing/gnome 47.
Built from source, ghosty opens fine, I have a single window, no split at all, but as soon as I hit the maximize button or resize just a tiny bit I get a segmentation fault.
This happens if I double click or run from command line (kitty and gnome terminal).
This happens with empty config or just a basic config like

gtk-single-instance=true
font-family="JetBrains Mono"

this is the log I get when I click the maximize button

(process:25246): Gtk-WARNING **: 18:13:58.826: GtkImage 0x3f658b90 reported baselines of minimum -2147483648 and natural -2147483648, but sizes of minimum 16 and natural 16. Baselines must be inside the widget size.
(process:25246): Gtk-WARNING **: 18:13:58.826: GtkImage 0x3f5221a0 reported baselines of minimum -2147483648 and natural -2147483648, but sizes of minimum 16 and natural 16. Baselines must be inside the widget size.
(process:25246): Gtk-WARNING **: 18:13:58.826: GtkImage 0x3f005590 reported baselines of minimum -2147483648 and natural -2147483648, but sizes of minimum 16 and natural 16. Baselines must be inside the widget size.
(process:25246): GLib-GObject-CRITICAL **: 18:13:58.859: g_object_ref: assertion 'G_IS_OBJECT (object)' failed
(process:25246): GLib-GObject-CRITICAL **: 18:13:58.859: g_object_ref: assertion 'G_IS_OBJECT (object)' failed
Segmentation fault

this is my setup

Dump report (added .txt at the end to be able to upload)

ffe49970-3ed7-4c11-375c-ff3876f77766.ghosttycrash.txt

My environment variables
env.txt

Is there anything else I can help with @mitchellh ?

[dariogriffo]

This definitely has to do with the build. If I download the ubuntu generate package from here resizing works fine...
I will try to create a debian package with debian dependencies and generate a .deb package!
@kjughx

[awsms]

This definitely has to do with the build.

This happens with my "own build", made with the ghostty-git package from the AUR.
https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=ghostty-git

edit: logs of a debug build before it crashes, https://0bin.net/paste/fL6wokg2#AyRKeuyEFtD-exBxWVoG0mlHS8WesjXRzp2r4CBoaYd

[kjughx]

I have a crash report if that helps?

[mitchellh]

If you have a stack trace that could help.

[pbsds]

My case might be related: ghostty for me becomes unstable and prone to crash if I have multiple active tabs open while disconnecting the monitor on my desktop (to preserve power).
When i reconnect the display i see the windows have been resized and the buffers are often corrupted before i force a redraw with reset. Sometimes the corruption lies dormant in a tab for hours, and once I select the tab ghostty freezes and crashes.

dump: https://gist.github.com/pbsds/ce21dbb112ca6376c5c97a3cf3143835 (no debug symbols though...)

Environment: nixos unstable (25.05), gnome 47 running on X11 with a NVIDIA GeForce GTX 1080 on linux 6.6.71 with opengl 4.6.0 NVIDIA 565.77.

Of interest I used scrollback-limit = 100000000 and gtk-wide-tabs = false

[flavorjones]

This matches my experience very closely. Ghostty has been stable for me except for when I connect/disconnect external monitors and resizing happens. I have not been able to get a core dump. I'm not sure how to help debug/diagnose this, but I'm willing to put in the work if someone has suggestions.

My environment is Ubuntu 24.10, kernel 6.11.0, Gnome 47, Wayland, with Intel graphics (RPL-P)

I've not set values for scrollback-limit or gtk-wide-tabs.

[lucaato]

I have the same problem, and I'm able to reproduce it when running the ghostty-debug package from the flake in the master. More info at this gist. The crash happens after aggressively resizing the splits in my experience, and this is also how i reproduced it.

[mitchellh]

Wow this is super helpful. Thank you. It's still missing some crucial data to make this trivial to figure out but this is a huge step forward to getting closer. Based on your gist, that terminal is in some weird state that is triggering an assertion, but it's unclear how it got there yet. Still, we can maybe think through this now.

[lucaato]

I'm not a zig dev so I can't be of much help by looking at the code. I've added to the same gist a file that contains a complete log, from ghostty startup to crash, which was not present before. Hope it helps in the resolution. If there's something I can do to ease the resolution I'd be glad to help.

[lucaato]

I updated the Ghostty flake to the version below, and the issue seems resolved, at least is way better than before. The crash didn't occur in the last few days. I don't know what changed, but updating could help some people.

├───ghostty: github:ghostty-org/ghostty/0eebebb75e6c5db9ab7947508ee0d8346d628339?narHash=sha256-Y2sj7S8bmqTaUB5dIfOvg08MVxf%2BIujS0GssRXbGfpE%3D (2025-05-11 21:45:30)
│   ├───flake-compat: github:edolstra/flake-compat/ff81ac966bb2cae68946d5ed5fc4994f96d0ffec?narHash=sha256-NeCCThCEP3eCl2l/%2B27kNNK7QrwZB1IJCrXfrbv5oqU%3D (2024-12-04 16:08:25)
│   ├───flake-utils: github:numtide/flake-utils/11707dc2f618dd54ca8739b309ec4fc024de578b?narHash=sha256-l0KFg5HjrsfsO/JpG%2Br7fRrqm12kzFHyUHqHCVpMMbI%3D (2024-11-13 21:27:16)
│   │   └───systems: github:nix-systems/default/da67096a3b9bf56a91d16901293e51ba5b49a27e?narHash=sha256-Vy1rq5AaRuLzOxct8nz4T6wlgyUR7zLU309k9mBC768%3D (2023-04-09 08:27:08)
│   ├───nixpkgs-stable: github:nixos/nixpkgs/da4b122f63095ca1199bd4d526f9e26426697689?narHash=sha256-nlIfTsTrMSksEJc1f7YexXiPVuzD1gOfeN1ggwZyUoc%3D (2025-03-14 22:42:37)
│   ├───nixpkgs-unstable: github:nixos/nixpkgs/573c650e8a14b2faa0041645ab18aed7e60f0c9a?narHash=sha256-4thdbnP6dlbdq%2BqZWTsm4ffAwoS8Tiq1YResB%2BRP6WE%3D (2025-03-13 11:38:39)
│   ├───zig: github:mitchellh/zig-overlay/0b14285e283f5a747f372fb2931835dd937c4383?narHash=sha256-aeopo%2BaXg5I2IksOPFN79usw7AeimH1%2BtjfuMzJHFdk%3D (2025-03-13 00:31:41)
│   │   ├───flake-compat follows input 'ghostty'
│   │   ├───flake-utils follows input 'ghostty/flake-utils'
│   │   └───nixpkgs follows input 'ghostty/nixpkgs-stable'
│   └───zon2nix: github:jcollie/zon2nix/56c159be489cc6c0e73c3930bd908ddc6fe89613?narHash=sha256-LhidlyEA9MP8jGe1rEnyjGFCzLLgCdDpYeWggibayr0%3D (2025-03-16 05:59:31)
│       ├───flake-utils follows input 'ghostty/flake-utils'
│       └───nixpkgs follows input 'ghostty/nixpkgs-unstable'

[rmunn]

I believe the intermittent crash-on-resize I've experienced has been this as well. I'm using the ghostty package from Arch, current version 1.1.3-1. The behavior I see is that most of the time, I can resize the ghostty terminal with no problem. But when there's a lot of text in the terminal (such as an active NeoVim session) and/or several tabs (I always have multiple Ghostty tabs running, so there is pretty much never just one tab) then I will sometimes see one of these two things:

• Stretched or compressed fonts, briefly, as if Ghostty is drawing 80 columns of text into a suddenly 146-column window (then the window figures out the correct column and row count for its new size and redraws correctly)
• A window that's displaying the wrong contents briefly, then a crash.

For the latter, what happened in the crash I just now saw is a terminal that showed only half the screen filled (I had two windows open in a vertical-split configuration, each sharing half the screen width, then I maximized the Ghostty terminal so its width doubled. It briefly showed the scrollback text wrapping at half the window width, then crashed after about three-quarters of a second).

This is happening maybe once every couple of weeks or so. I would love to submit a crash report, but as far as I can tell my Ghostty installation doesn't seem to be producing them. Please let me know how I can enable crash reporting (do I need to download a debug build for that?) and then I'll be able to contribute crash reports.

[00-kat]

The crash reports weren't useful on Linux so they've been disabled now. You can run a debug build to see a trace in the logs though, but debug builds are very slow so don't use it unless you can consistently reproduce it.

[rmunn]

I haven't been able to reproduce the crashes consistently, so I guess I'll just give verbal reports of what I was doing and/or running each time I experience once and maybe I'll eventually find a pattern that would let me reproduce the issue. But please let me know if there's any way I can provide more data short of running a debug build.

[alaviss]

Thanks to @lucaato stack trace, here's a stress test that reproduces the assertion when run on Debug tip builds.

1. cat this file to your terminal. This contains 1000 👨‍👨‍👦‍👦 characters. Note that this portion is quite slow due to excessive debug prints.
2. Run cat (without arguments). This prevents whatever shell you're using from reacting to SIGWINCH and redraw.
3. Split left/right. (Your terminal might crash here)
4. Run step 1-2 on the split.
5. Resize the splits rapidly. This more or less guarantees the crash, or a rendering bug where all emojis disappears.

Stack trace from GDB:

#0  __pthread_kill_implementation (threadid=<optimized out>, signo=signo@entry=6, no_tid=no_tid@entry=0)
    at pthread_kill.c:44
#1  0x00007ffff685fe23 in __pthread_kill_internal (threadid=<optimized out>, signo=6) at pthread_kill.c:78
#2  0x00007ffff680708e in __GI_raise (sig=sig@entry=6) at ../sysdeps/posix/raise.c:26
#3  0x00007ffff67ee882 in __GI_abort () at abort.c:79
#4  0x000055555785bcb1 in posix.abort () at /usr/lib/sdk/ziglang/lib/std/posix.zig:710
#5  0x00005555578572ad in debug.defaultPanic (msg=..., first_trace_addr=...)
    at /usr/lib/sdk/ziglang/lib/std/debug.zig:704
#6  0x000055555785ee59 in debug.FullPanic((function 'defaultPanic')).reachedUnreachable ()
    at /usr/lib/sdk/ziglang/lib/std/debug.zig:63
#7  0x0000555557859a6e in debug.assert (ok=false) at /usr/lib/sdk/ziglang/lib/std/debug.zig:550
#8  0x0000555557e000cd in terminal.page.Page.setGraphemes
    (self=0x7fffcd5c09c8, row=0x7fffe9244008, cell=0x7fffe9246038, cps=...)
    at /run/build/ghostty/src/terminal/page.zig:1322
#9  0x0000555557e4e43c in terminal.PageList.ReflowCursor.reflowRow (self=0x7fff56ff85f8, list=0x7fffea9db4b0, row=...)
    at /run/build/ghostty/src/terminal/PageList.zig:1018
#10 0x0000555557e4fe43 in terminal.PageList.resizeCols (self=0x7fffea9db4b0, cols=49, cursor=...)
    at /run/build/ghostty/src/terminal/PageList.zig:727
#11 0x0000555557e50adb in terminal.PageList.resize (self=0x7fffea9db4b0, opts=...)
    at /run/build/ghostty/src/terminal/PageList.zig:654
#12 0x0000555557e50dc9 in terminal.Screen.resizeInternal (self=0x7fffea9db4a0, cols=49, rows=59, reflow=true)
    at /run/build/ghostty/src/terminal/Screen.zig:1616
#13 0x0000555557e51734 in terminal.Screen.resize (self=0x7fffea9db4a0, cols=49, rows=59)
    at /run/build/ghostty/src/terminal/Screen.zig:1544
#14 0x0000555557e51b5a in terminal.Terminal.resize (self=0x7fffea9db4a0, alloc=..., cols=49, rows=59)
    at /run/build/ghostty/src/terminal/Terminal.zig:2467
#15 0x0000555557e52385 in termio.Termio.resize (self=0x7fffea9daeb0, td=0x7fff56ff9ec8, size=...)
    at /run/build/ghostty/src/termio/Termio.zig:370
#16 0x0000555557e527d0 in termio.Thread.coalesceCallback (cb_=0x7fff56ff9eb8, r=0)
    at /run/build/ghostty/src/termio/Thread.zig:384
#17 0x0000555557ee1172 in watcher.timer.TimerDynamic(dynamic.Xev(&.{ .io_uring, .epoll }[0..2])).reset__anon_464600__struct_480776.callback ()
    at /run/build/ghostty/vendor/p/libxev-0.0.0-86vtc-ziEgDbLP0vihUn1MhsxNKY4GJEga6BEr7oyHpz/src/watcher/timer.zig:345
#18 watcher.timer.TimerLoop(main.Xev(.io_uring,backend.io_uring)).reset__anon_480782__struct_490394.callback
    (ud=0x7fff56ff9eb8, l_inner=0x7fffea9dcc78, c_inner=0x7fffea9dcf68, r=...)
    at /run/build/ghostty/vendor/p/libxev-0.0.0-86vtc-ziEgDbLP0vihUn1MhsxNKY4GJEga6BEr7oyHpz/src/watcher/timer.zig:122
#19 0x0000555557d277ff in backend.io_uring.Completion.invoke (self=0x7fffea9dcf68, loop=0x7fffea9dcc78, res=-62)
    at /run/build/ghostty/vendor/p/libxev-0.0.0-86vtc-ziEgDbLP0vihUn1MhsxNKY4GJEga6BEr7oyHpz/src/backend/io_uring.zig:800
#20 0x0000555557d2a429 in backend.io_uring.Loop.tick___anon_404329 (self=0x7fffea9dcc78)
    at /run/build/ghostty/vendor/p/libxev-0.0.0-86vtc-ziEgDbLP0vihUn1MhsxNKY4GJEga6BEr7oyHpz/src/backend/io_uring.zig:189
#21 0x0000555557d2a51e in backend.io_uring.Loop.run (self=0x7fffea9dcc78, mode=until_done)
    at /run/build/ghostty/vendor/p/libxev-0.0.0-86vtc-ziEgDbLP0vihUn1MhsxNKY4GJEga6BEr7oyHpz/src/backend/io_uring.zig:80
#22 0x0000555557d2efc4 in dynamic.Xev(&.{ .io_uring, .epoll }[0..2]).Loop.run (self=0x7fffea9dcc78, mode=until_done)
    at /run/build/ghostty/vendor/p/libxev-0.0.0-86vtc-ziEgDbLP0vihUn1MhsxNKY4GJEga6BEr7oyHpz/src/dynamic.zig:172
#23 0x0000555557d35122 in termio.Thread.threadMain_ (self=0x7fffea9dcc68, io=0x7fffea9daeb0)
    at /run/build/ghostty/src/termio/Thread.zig:237
#24 0x0000555557cca1ef in termio.Thread.threadMain (self=0x7fffea9dcc68, io=0x7fffea9daeb0)
    at /run/build/ghostty/src/termio/Thread.zig:123
#25 0x0000555557c6d342 in Thread.callFn__anon_383021 (args=Python Exception <class 'gdb.MemoryError'>: Cannot access memory at address 0x0

   #26 0x0000555557beca35 in Thread.PosixThreadImpl.spawn__anon_222451.Instance.entryFn (raw_arg=0x55555a0ba1f0)
    at /usr/lib/sdk/ziglang/lib/std/Thread.zig:757
#27 0x00007ffff685dce1 in start_thread (arg=<optimized out>) at pthread_create.c:447
#28 0x00007ffff68e27d4 in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:100

Some final log lines since the entire thing is too big:

debug(terminal): c=200D grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=1F468 grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=200D grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=1F466 grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=200D grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=1F466 grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=200D grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=1F468 grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=200D grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=1F466 grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=200D grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=1F466 grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=200D grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=1F468 grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=200D grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=1F466 grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=200D grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=1F466 grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=200D grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=1F468 grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=200D grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=1F466 grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=200D grapheme attach to left=2 primary_cp=1F468
debug(terminal): c=1F466 grapheme attach to left=2 primary_cp=1F468
debug(io_thread): mailbox message=termio.message.Message{ .focused = false }
debug(app): focus event focused=false
info(grid): reallocating GPU buffer old=360 new=817
debug(app): mailbox message=redraw_surface
debug(renderer_thread): mailbox message=renderer.message.Message{ .reset_cursor_blink = void }
debug(renderer_thread): mailbox message=renderer.message.Message{ .focus = false }
debug(renderer_thread): mailbox message=renderer.message.Message{ .reset_cursor_blink = void }
debug(renderer_thread): mailbox message=renderer.message.Message{ .reset_cursor_blink = void }
debug(renderer_thread): mailbox message=renderer.message.Message{ .reset_cursor_blink = void }
debug(renderer_thread): mailbox message=renderer.message.Message{ .reset_cursor_blink = void }
debug(renderer_thread): mailbox message=renderer.message.Message{ .reset_cursor_blink = void }
debug(renderer_thread): mailbox message=renderer.message.Message{ .reset_cursor_blink = void }
debug(renderer_thread): mailbox message=renderer.message.Message{ .reset_cursor_blink = void }
info(grid): reallocating GPU buffer old=817 new=1228
debug(app): mailbox message=redraw_surface
debug(app): mailbox message=redraw_surface
debug(app): mailbox message=redraw_surface
debug(app): mailbox message=redraw_surface
debug(app): mailbox message=redraw_surface
debug(renderer_thread): mailbox message=renderer.message.Message{ .focus = true }
debug(app): focus event focused=true
debug(io_thread): mailbox message=termio.message.Message{ .focused = true }
Creating EGL context version 3.3 (debug:no, forward:no, legacy:no, es:no)
[New Thread 0x7fff8a7fc6c0 (LWP 58)]
[New Thread 0x7fff877f66c0 (LWP 59)]
[New Thread 0x7fff8b7fe6c0 (LWP 60)]
OpenGL version: 4.6 (core)
GLSL version: 4.60
Max texture size: 16384
Enabled features (use GDK_GL_DISABLE env var to disable):
    debug: ✓
    base-instance: ✓
    buffer-storage: ✓
debug(app): focus event focused=false
debug(app): focus event focused=true
debug(app): focus event focused=false
debug(app): focus event focused=true
debug(app): mailbox message=redraw_surface
debug(app): mailbox message=redraw_surface
debug(app): mailbox message=redraw_surface
debug(io_thread): mailbox message=termio.message.Message{ .focused = false }
debug(renderer_thread): mailbox message=renderer.message.Message{ .focus = false }
debug(app): focus event focused=false
debug(io_thread): mailbox message=termio.message.Message{ .focused = true }
debug(app): focus event focused=true
debug(app): mailbox message=redraw_surface
debug(renderer_thread): mailbox message=renderer.message.Message{ .focus = true }
debug(app): mailbox message=redraw_surface
debug(io_thread): mailbox message=termio.message.Message{ .write_small = termio.message.MessageData(u8,38).Small{ .data = { 99, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170 }, .len = 1 } }
debug(renderer_thread): mailbox message=renderer.message.Message{ .reset_cursor_blink = void }
debug(app): mailbox message=redraw_surface
debug(app): mailbox message=redraw_surface
debug(io_thread): mailbox message=termio.message.Message{ .write_small = termio.message.MessageData(u8,38).Small{ .data = { 97, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170 }, .len = 1 } }
debug(io_thread): mailbox message=termio.message.Message{ .write_small = termio.message.MessageData(u8,38).Small{ .data = { 116, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170 }, .len = 1 } }
debug(renderer_thread): mailbox message=renderer.message.Message{ .reset_cursor_blink = void }
debug(app): mailbox message=redraw_surface
debug(app): mailbox message=redraw_surface
debug(io_thread): mailbox message=termio.message.Message{ .write_small = termio.message.MessageData(u8,38).Small{ .data = { 13, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170, 170 }, .len = 1 } }
debug(io_exec): termios change mode=pty.Mode{ .canonical = true, .echo = true }
debug(renderer_thread): mailbox message=renderer.message.Message{ .reset_cursor_blink = void }
debug(app): mailbox message=redraw_surface
debug(app): mailbox message=redraw_surface
debug(app): focus event focused=false
debug(app): focus event focused=true
debug(app): focus event focused=false
debug(app): focus event focused=true
Creating EGL context version 3.3 (debug:no, forward:no, legacy:no, es:no)
[New Thread 0x7fff87ff76c0 (LWP 61)]
[New Thread 0x7fffe90136c0 (LWP 62)]
[New Thread 0x7fff8bfff6c0 (LWP 63)]
OpenGL version: 4.6 (core)
GLSL version: 4.60
Max texture size: 16384
Enabled features (use GDK_GL_DISABLE env var to disable):
    debug: ✓
    base-instance: ✓
    buffer-storage: ✓
debug(app): focus event focused=false
debug(app): focus event focused=true
debug(app): focus event focused=false
debug(app): focus event focused=true
Creating EGL context version 3.3 (debug:no, forward:no, legacy:no, es:no)
[New Thread 0x7fff8affd6c0 (LWP 64)]
[New Thread 0x7fff89ffb6c0 (LWP 65)]
[New Thread 0x7fff897fa6c0 (LWP 66)]
OpenGL version: 4.6 (core)
GLSL version: 4.60
Max texture size: 16384
Enabled features (use GDK_GL_DISABLE env var to disable):
    debug: ✓
    base-instance: ✓
    buffer-storage: ✓
debug(app): focus event focused=false
debug(app): focus event focused=true
debug(app): focus event focused=false
debug(app): focus event focused=true
debug(app): focus event focused=false
debug(app): focus event focused=true
debug(app): focus event focused=false
debug(app): focus event focused=true
debug(app): focus event focused=false
debug(app): focus event focused=true
debug(app): focus event focused=false
debug(app): focus event focused=true
debug(gtk_surface): gl surface unrealized
debug(opengl): program destroyed id=6
debug(opengl): program destroyed id=3
Destroying EGL context
[Thread 0x7fff857f26c0 (LWP 49) exited]
[Thread 0x7fff85ff36c0 (LWP 48) exited]
[Thread 0x7fff867f46c0 (LWP 47) exited]
Destroying EGL context
[Thread 0x7fff8bfff6c0 (LWP 63) exited]
[Thread 0x7fffe90136c0 (LWP 62) exited]
[Thread 0x7fff87ff76c0 (LWP 61) exited]
Destroying EGL context
[Thread 0x7fff897fa6c0 (LWP 66) exited]
[Thread 0x7fff89ffb6c0 (LWP 65) exited]
[Thread 0x7fff8affd6c0 (LWP 64) exited]
Creating EGL context version 3.3 (debug:no, forward:no, legacy:no, es:no)
[New Thread 0x7fff8affd6c0 (LWP 67)]
[New Thread 0x7fff89ffb6c0 (LWP 68)]
[New Thread 0x7fff897fa6c0 (LWP 69)]
debug(gtk_surface): gl surface realized
Destroying EGL context
[Thread 0x7fff8b7fe6c0 (LWP 60) exited]
[Thread 0x7fff877f66c0 (LWP 59) exited]
[Thread 0x7fff8a7fc6c0 (LWP 58) exited]
OpenGL version: 4.6 (core)
GLSL version: 4.60
Max texture size: 16384
Enabled features (use GDK_GL_DISABLE env var to disable):
    debug: ✓
    base-instance: ✓
    buffer-storage: ✓
info(grid): loaded OpenGL 4.6
debug(surface): xscale=1e0 yscale=1e0 xdpi=9.6e1 ydpi=9.6e1
debug(font_shared_grid_set): found cached grid for font config
debug(opengl): shader created id=1
debug(opengl): shader created id=2
debug(opengl): program created id=3
debug(opengl): program linked id=3
debug(opengl): shader destroyed id=2
debug(opengl): shader destroyed id=1
debug(opengl): shader created id=4
debug(opengl): shader created id=5
debug(opengl): program created id=6
debug(opengl): program linked id=6
debug(opengl): shader destroyed id=5
debug(opengl): shader destroyed id=4
info(io_exec): found Ghostty resources dir: /app/share/ghostty
info(io_exec): shell integration automatically injected shell=termio.shell_integration.Shell.bash
warning(gtk): unimplemented action=apprt.action.Action.Key.cell_size
[New Thread 0x7fff86ff56c0 (LWP 70)]
debug(renderer_thread): starting renderer thread
[New Thread 0x7fff85ff46c0 (LWP 71)]
debug(surface): set font size size=1.2e1
debug(font_shared_grid_set): found cached grid for font config
warning(gtk): unimplemented action=apprt.action.Action.Key.cell_size
debug(io_exec): starting command command={ /bin/bash, --posix }
[New Thread 0x7fff3bfff6c0 (LWP 72)]
Creating EGL context version 3.3 (debug:no, forward:no, legacy:no, es:no)
[New Thread 0x7fff8a7fc6c0 (LWP 73)]
[New Thread 0x7fff877f66c0 (LWP 74)]
[New Thread 0x7fff8b7fe6c0 (LWP 75)]
debug(gtk_surface): gl surface realized
OpenGL version: 4.6 (core)
GLSL version: 4.60
Max texture size: 16384
Enabled features (use GDK_GL_DISABLE env var to disable):
    debug: ✓
    base-instance: ✓
    buffer-storage: ✓
debug(opengl): shader created id=7
debug(opengl): shader created id=8
debug(opengl): program created id=25
debug(renderer_thread): mailbox message=renderer.message.Message{ .font_grid = renderer.message.Message__struct_156186{ .grid = font.SharedGrid{ .codepoints = hash_map.HashMapUnmanaged(font.SharedGrid.CodepointKey,?font.Collection.Index,hash_map.AutoContext(font.SharedGrid.CodepointKey),80){ ... }, .glyphs = hash_map.HashMapUnmanaged(font.SharedGrid.GlyphKey,font.SharedGrid.Render,hash_map.AutoContext(font.SharedGrid.GlyphKey),80){ ... }, .atlas_grayscale = font.Atlas{ ... }, .atlas_color = font.Atlas{ ... }, .resolver = font.CodepointResolver{ ... }, .metrics = font.Metrics{ ... }, .lock = Thread.RwLock{ ... } }, .set = font.SharedGridSet{ .alloc = mem.Allocator{ ... }, .map = hash_map.HashMapUnmanaged(font.SharedGridSet.Key,font.SharedGridSet.ReffedGrid,font.SharedGridSet.Map__struct_150741,80){ ... }, .font_lib = font.library.FreetypeLibrary{ ... }, .font_discover = font.discovery.Fontconfig{ ... }, .lock = Thread.Mutex{ ... } }, .new_key = font.SharedGridSet.Key{ .arena = heap.arena_allocator.ArenaAllocator{ ... }, .descriptors = { ... }, .style_offsets = { ... }, .codepoint_map = font.CodepointMap{ ... }, .metric_modifiers = hash_map.HashMapUnmanaged(font.Metrics.Key__enum_151010,font.Metrics.Modifier,hash_map.AutoContext(font.Metrics.Key__enum_151010),80){ ... }, .font_size = font.face.DesiredSize{ ... }, .freetype_load_flags = config.Config.FreetypeLoadFlags{ ... } }, .old_key = font.SharedGridSet.Key{ .arena = heap.arena_allocator.ArenaAllocator{ ... }, .descriptors = { ... }, .style_offsets = { ... }, .codepoint_map = font.CodepointMap{ ... }, .metric_modifiers = hash_map.HashMapUnmanaged(font.Metrics.Key__enum_151010,font.Metrics.Modifier,hash_map.AutoContext(font.Metrics.Key__enum_151010),80){ ... }, .font_size = font.face.DesiredSize{ ... }, .freetype_load_flags = config.Config.FreetypeLoadFlags{ ... } } } }
debug(opengl): program linked id=25
debug(opengl): shader destroyed id=8
debug(opengl): shader destroyed id=7
debug(opengl): shader created id=26
debug(opengl): shader created id=27
debug(opengl): program created id=28
debug(opengl): program linked id=28
debug(opengl): shader destroyed id=27
debug(opengl): shader destroyed id=26
debug(renderer_thread): mailbox message=renderer.message.Message{ .focus = false }
debug(io_thread): mailbox message=termio.message.Message{ .focused = false }
debug(app): mailbox message=redraw_surface
debug(gtk_surface): gl resize width=500 height=1310 scale=1 window_scale=1
debug(gtk_surface): gl resize width=499 height=1310 scale=1 window_scale=1
info(grid): reallocating GPU buffer old=0 new=1228
debug(app): mailbox message=redraw_surface
debug(io_thread): mailbox message=termio.message.Message{ .resize = renderer.size.Size{ .screen = renderer.size.ScreenSize{ .width = 499, .height = 1310 }, .cell = renderer.size.CellSize{ .width = 10, .height = 22 }, .padding = renderer.size.Padding{ .top = 2, .bottom = 2, .right = 2, .left = 2 } } }
thread 51 panic: reached unreachable code
Unwind information for `:0x55555787cbfd` was not available, trace may be incomplete

???:?:?: 0x555557859a6d in ??? ()
???:?:?: 0x555557e000cc in ??? ()
???:?:?: 0x555557e4e43b in ??? ()
???:?:?: 0x555557e4fe42 in ??? ()
???:?:?: 0x555557e50ada in ??? ()
???:?:?: 0x555557e50dc8 in ??? ()
???:?:?: 0x555557e51733 in ??? ()
???:?:?: 0x555557e51b59 in ??? ()
???:?:?: 0x555557e52384 in ??? ()
???:?:?: 0x555557e527cf in ??? ()
???:?:?: 0x555557ee1171 in ??? ()
???:?:?: 0x555557d277fe in ??? ()
???:?:?: 0x555557d2a428 in ??? ()
???:?:?: 0x555557d2a51d in ??? ()
???:?:?: 0x555557d2efc3 in ??? ()
???:?:?: 0x555557d35121 in ??? ()
???:?:?: 0x555557cca1ee in ??? ()
???:?:?: 0x555557c6d341 in ??? ()
???:?:?: 0x555557beca34 in ??? ()
???:?:?: 0x7ffff685dce0 in ??? (libc.so.6)
???:?:?: 0x7ffff68e27d3 in ??? (libc.so.6)

Thread 38 "io" received signal SIGABRT, Aborted.

[mitchellh]

Awesome! This is excellent. There's clearly some kind of accounting issues when it comes to graphemes and resizing. It's unclear exactly what that is but this should be enough to figure it out. I'm going to make an issue.

[mitchellh]

Issue #7536

[rmunn]

I just experienced another crash after resizing while running Ghostty tip with the bugfix included, so the cause of my crashes might have been different from #7536. I think I'll switch to a debug build for a while so I can get a log file or crash report to submit. If I do, I'll probably open a new discussion so that this one can remain focused on issue #7536.

[rmunn]

New discussion (with crash log) opened in #7977.

[rmunn]

#7977 now has a transcript file that you can cat to the terminal to reproduce the crash (reproduces consistently in debug build, does not crash in release build). File attached to #7977 (comment)

[jcollie]

Closing this as it's become a collection point for various unrelated crashes. Please open up a new discussion if you have a similar problems.