Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[xcb] Unknown sequence number while processing queue SIGABRT #54

Closed
CamilleScholtz opened this issue Oct 6, 2019 · 21 comments
Closed

Comments

@CamilleScholtz
Copy link

No idea why this happens, freshly compiled and no changed made, error:

[xcb] Unknown sequence number while processing queue
[xcb] Most likely this is a multi-threaded client and XInitThreads has not been called
[xcb] Aborting, sorry about that.
berry: xcb_io.c:263: poll_for_event: Assertion `!xcb_xlib_threads_sequence_lost' failed.
fish: Job 3, '~/tmp/berry/berry' terminated by signal SIGABRT (Abort)
@JLErvin
Copy link
Owner

JLErvin commented Oct 6, 2019

This is a curious issue since berry uses the xlib library, not xcb. Still, it could be something on our end - although I'm more convinced it's some difficult to track down race condition.

Could you post your log file/try to recreate the error so I can do some more thorough investigation?

@dylanaraps
Copy link

On my machine libX11 (xlib) depends on libxcb (I'm not sure if this is always the case). I've come across this issue too and I was also unable to fix it.

The steps to reproduce it on my machine are simply:

  • Compile berry (make)
  • Launch berry.
  • Above error.

I'm running with musl libc for reference.

@dylanaraps
Copy link

libX11 always requires libxcb:

# Always required
X11_REQUIRES='xproto >= 7.0.17 xextproto xtrans xcb >= 1.11.1'
X11_EXTRA_DEPS="xcb >= 1.11.1"

From: https://github.com/mirror/libX11/blob/master/configure.ac#L80-L82

@JLErvin
Copy link
Owner

JLErvin commented Oct 6, 2019

Yes that's true - I just meant that there is no explicit xcb code used inside of berry.

Nonetheless, I'll setup a VM with some musl-based distro and try to debug this.

@dylanaraps
Copy link

Note: I'm unsure if @onodera-punpun is using musl also. This may be reproducible on glibc too.

@dylanaraps
Copy link

dylanaraps commented Oct 6, 2019

Not sure if this is of any use to you as it's related to a rust library, however it looks like the same issue: rust-windowing/winit#458

The fix is described as:

Fixed race conditions when creating an EventsLoop on X11, most commonly manifesting as "[xcb] Unknown sequence number while processing queue".

@JLErvin
Copy link
Owner

JLErvin commented Oct 6, 2019

@dylanaraps I saw that as well but I'm not sure if it's applicable to our problem yet.

@dylanaraps
Copy link

@CamilleScholtz
Copy link
Author

CamilleScholtz commented Oct 7, 2019

I'm using glibc, although I do use a source distro, so it could possible be the result of a CFLAG or kernel option I set. However I never saw this error before with any other wm or X app.

@JLErvin
Copy link
Owner

JLErvin commented Oct 7, 2019

@onodera-punpun I've haven't seen this issue elsewhere either. Can you post your USE flags (or equivalent) so I can try to recreate the issue on my end?

@CamilleScholtz
Copy link
Author

Oh wait, I didn't build this package with my package manager, so in fact I just used the default CFLAGS as defined in config.mk...

@JLErvin
Copy link
Owner

JLErvin commented Oct 10, 2019

I added a potential fix. Since I still haven't been able to recreate this on my end could someone try the development branch, raspberry, and see if this fixes the issue? @onodera-punpun

@CamilleScholtz
Copy link
Author

Sadly didn't fix the issue, exact same error

@polzon
Copy link

polzon commented Oct 18, 2019

I'm on debian 10 stable, and also am getting the same error when I run startx /usr/local/bin/berry

edit: still get the same error on the raspberry branch.

The xorg server will start, and I'll be able to see a blank desktop, but as soon as I left or right click it will crash back to console and give me OP's same error.

Also just tested it on my desktop which runs a near-fresh install of Debian 10 and it doesn't seem to have this issue.

@JLErvin
Copy link
Owner

JLErvin commented Oct 29, 2019

@dylanaraps can you post your global compiler options for KISS?

@BartolomeyKant
Copy link

I have the same issue on Void Linux. I install berry from void linux repository. Also i tried build from sources.
in my ~/.xinintrc i add

berry &
exec xterm

so after startx i have black screen with xterm but berry not working. If i restart berry from terminal with -d option it says same errors like @onodera-punpun posted.

@BartolomeyKant
Copy link

I trying to debug this error.
berry_log.txt

Testing on manjaro, berry builded from source.
At start there is no errors (line 14). But on first event (mouse click or start new application) it crushes (after line 14).

@BartolomeyKant
Copy link

I think I know how to repeate this.
I add x permission to .config/berry/autostart and now it working
if I remove x permission it's not working with same error

@JLErvin
Copy link
Owner

JLErvin commented Feb 19, 2020

@NotEviLL yes! oh my goodness thank you I've been unable to recreate this for months. For now, make it executable and I'll try to find a more universal fix :)

@JLErvin
Copy link
Owner

JLErvin commented Feb 19, 2020

@NotEviLL can you try using the development branch, grape, and see if this works without autostart being executable. branch here

@BartolomeyKant
Copy link

Yes, it's working

@JLErvin JLErvin closed this as completed Feb 19, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

5 participants