Re: UPDATE: games/minetest to 5.9.1

#0 0x000000b4fccbf9e4 in XQueryExtension () from /usr/X11R6/lib/libX11.so.18.2
#1 0x000000b4fccb1706 in XInitExtension () from /usr/X11R6/lib/libX11.so.18.2
#2 0x000000b47a594759 in __glXInitialize (dpy=0x0)
at /usr/xenocara/lib/mesa/mk/libGL/../../src/glx/glxext.c:852
#3 0x000000b47a5950ed in __glXSetupForCommand (dpy=0x0)
at /usr/xenocara/lib/mesa/mk/libGL/../../src/glx/glxext.c:993
#4 0x000000b47a5a7777 in protocolDestroyDrawable (dpy=0xb47a571980 <__glXExtensionName>,
drawable=775146641792, glxCode=3)
at /usr/xenocara/lib/mesa/mk/libGL/../../src/glx/glx_pbuffer.c:399
#5 0x000000b47a5a72eb in DestroyDrawable (dpy=0xb47a571980 <__glXExtensionName>,
drawable=775146641792, glxCode=32)
at /usr/xenocara/lib/mesa/mk/libGL/../../src/glx/glx_pbuffer.c:505
#6 glXDestroyWindow (dpy=0xb47a571980 <__glXExtensionName>, win=775146641792)
at /usr/xenocara/lib/mesa/mk/libGL/../../src/glx/glx_pbuffer.c:921
#7 0x000000b2114d6087 in irr::CIrrDeviceLinux::~CIrrDeviceLinux() ()
#8 0x000000b2114d6364 in virtual thunk to irr::CIrrDeviceLinux::~CIrrDeviceLinux() ()
#9 0x000000b21167e71f in RenderingEngine::~RenderingEngine() ()
#10 0x000000b2115c157e in ClientLauncher::~ClientLauncher() ()
#11 0x000000b2119eeaa3 in main ()
On Wed, 16 Oct 2024 20:13:45 -0500
izzy Meyer <izder456@disroot.org> wrote:

[...]

> Also, minetest_game (the de-facto game for this engine), became
> rolling release in this 5.9.x lineage. The engine itself can
> automatically download the game, so it doesn't make much sense to
> package it in our port.
>
> Here's the changelog for 5.9.1:
>
> https://dev.minetest.net/Changelog#5.9.0_.E2.86.92_5.9.1
>
> Diff is attached.
>
> Any committer want to look this over?

I looked at it and have a few requests:

1. Please consider taking maintainer, as you've been most active for
this port and this makes it easier to have continuity.

2. make port-lib-depends-check:
Missing lib: atomic.3 (/usr/local/bin/minetestserver) (NOT REACHABLE)
Appears to be something from gcc-libs...? I didn't further investigate,
but seems odd for this to show up and should probably be
explained/accounted for.

3.Builds and runs fine with a short test, but I'm getting a segfault at
exit (attached). Is this new, and is it known upstream?

4. Lastly, while minor I would prefer harmonizing the use of tabs -
lang/luajit has 2 more tabs than the rest, and PKGNAME, DIST_TUPLE, and
CATEGORIES lines could get 1 more tab so that it's easier to read.