Re: GNU G-Golf 0.8.0-rc-3 available for testing

[pelzflorian (Florian Pelz)]

Hi David.  (gc-disable) was a smart idea, sadly it does not help, but at
least we know now.  It makes Valgrind errors disappear after
(gc-disable) was called.

scheme@(guile-user)> ,use (g-golf)
scheme@(guile-user)> (gc-disable)
scheme@(guile-user)> (chdir "/home/florian/src/g-golf/examples/gtk-4")
scheme@(guile-user)> (load "drawing-widget.scm")
scheme@(guile-user)> (main '("-d"))
;; g-application-run 
;;          application: #<<gtk-application> 7ff0b42834e0> 
;;                 argc: 0 
;;                 argv: () 
;; gtk-window-set-child 
;;               window: #<<gtk-application-window> 7ff0b42a0eb0> 
;;                child: #<<drawing-widget> 7ff0b42a0bf0> 
;; gtk-window-present 
;;               window: #<<gtk-application-window> 7ff0b42a0eb0> 
Segmentation fault
florian@florianhp ~/src/tcgocr [env]$ valgrind 2>&1 guile | tee valout
==7880== Memcheck, a memory error detector
==7880== Copyright (C) 2002-2022, and GNU GPL'd, by Julian Seward et al.
==7880== Using Valgrind-3.22.0 and LibVEX; rerun with -h for copyright info
==7880== Command: guile
==7880== 
==7880== Conditional jump or move depends on uninitialised value(s)
==7880==    at 0x49B5016: GC_push_all_eager (in 
/gnu/store/1w1r6r56z9lhg8ghcb7lxss6mkn7d5l1-libgc-8.2.2/lib/libgc.so.1.5.1)
==7880==    by 0x49AB38B: ??? (in 
/gnu/store/1w1r6r56z9lhg8ghcb7lxss6mkn7d5l1-libgc-8.2.2/lib/libgc.so.1.5.1)
[tons of GC errors…]
scheme@(guile-user)> (gc-disable)
[more GC errors…]
==7880==    by 0x48C9EDB: ??? (in 
/gnu/store/4gvgcfdiz67wv04ihqfa8pqwzsb0qpv5-guile-3.0.9/lib/libguile-3.0.so.1.6.0)
==7880==    by 0x492895E: ??? (in 
/gnu/store/4gvgcfdiz67wv04ihqfa8pqwzsb0qpv5-guile-3.0.9/lib/libguile-3.0.so.1.6.0)
==7880==    by 0x4933DB4: scm_call_n (in 
/gnu/store/4gvgcfdiz67wv04ihqfa8pqwzsb0qpv5-guile-3.0.9/lib/libguile-3.0.so.1.6.0)
==7880== 
scheme@(guile-user)> ,use (g-golf)
scheme@(guile-user)> (chdir "/home/florian/src/g-golf/examples/gtk-4")
scheme@(guile-user)> (load "drawing-widget.scm")
scheme@(guile-user)> (main '("-d"))
==7880== 
==7880== Process terminating with default action of signal 11 (SIGSEGV)
==7880==  Bad permissions for mapped region at address 0x94BC2D0
==7880==    at 0x94BC2D0: ???
==7880==    by 0xFEBB28A: ??? (in 
/gnu/store/qp9j03r0qszb61j6w786nvcqpd6f0rij-gtk-4.12.3/lib/libgtk-4.so.1.1200.3)
==7880==    by 0xFEBC189: gtk_widget_snapshot_child (in 
/gnu/store/qp9j03r0qszb61j6w786nvcqpd6f0rij-gtk-4.12.3/lib/libgtk-4.so.1.1200.3)
==7880==    by 0xFEA3FD3: ??? (in 
/gnu/store/qp9j03r0qszb61j6w786nvcqpd6f0rij-gtk-4.12.3/lib/libgtk-4.so.1.1200.3)
==7880==    by 0xFEBB0F7: ??? (in 
/gnu/store/qp9j03r0qszb61j6w786nvcqpd6f0rij-gtk-4.12.3/lib/libgtk-4.so.1.1200.3)
==7880==    by 0xFEBB28A: ??? (in 
/gnu/store/qp9j03r0qszb61j6w786nvcqpd6f0rij-gtk-4.12.3/lib/libgtk-4.so.1.1200.3)
==7880==    by 0xFEBB342: ??? (in 
/gnu/store/qp9j03r0qszb61j6w786nvcqpd6f0rij-gtk-4.12.3/lib/libgtk-4.so.1.1200.3)
==7880==    by 0xFEBB449: ??? (in 
/gnu/store/qp9j03r0qszb61j6w786nvcqpd6f0rij-gtk-4.12.3/lib/libgtk-4.so.1.1200.3)
==7880==    by 0xFECACA0: ??? (in 
/gnu/store/qp9j03r0qszb61j6w786nvcqpd6f0rij-gtk-4.12.3/lib/libgtk-4.so.1.1200.3)
==7880==    by 0x10100370: ??? (in 
/gnu/store/qp9j03r0qszb61j6w786nvcqpd6f0rij-gtk-4.12.3/lib/libgtk-4.so.1.1200.3)
==7880==    by 0xC0F05B8: ??? (in 
/gnu/store/kjdjarbkknf8vv9rc8vkkh1k1vrc57r0-glib-2.78.0/lib/libgobject-2.0.so.0.7800.0)
==7880==    by 0xC10352D: ??? (in 
/gnu/store/kjdjarbkknf8vv9rc8vkkh1k1vrc57r0-glib-2.78.0/lib/libgobject-2.0.so.0.7800.0)
==7880== 
==7880== HEAP SUMMARY:
==7880==     in use at exit: 36,575,081 bytes in 110,686 blocks
==7880==   total heap usage: 633,552 allocs, 522,866 frees, 87,881,755 bytes 
allocated
==7880== 
==7880== LEAK SUMMARY:
==7880==    definitely lost: 2,936 bytes in 15 blocks
==7880==    indirectly lost: 16,350 bytes in 682 blocks
==7880==      possibly lost: 19,128,578 bytes in 14,960 blocks
==7880==    still reachable: 17,316,057 bytes in 94,084 blocks
==7880==         suppressed: 0 bytes in 0 blocks
==7880== Rerun with --leak-check=full to see details of leaked memory
==7880== 
==7880== Use --track-origins=yes to see where uninitialised values come from
==7880== For lists of detected and suppressed errors, rerun with: -s
==7880== ERROR SUMMARY: 9656 errors from 72 contexts (suppressed: 0 from 0)
florian@florianhp ~/src/tcgocr [env]$ 



David Pirotte <david@altosw.be> writes:
> No problem. I am also interested in fixing this bug

:)

> Imo, you should locally patch your guix image so it has one gdk-pixbuf
> lib installed  keep the one that has the debug symbols

The one gdk-pixbuf problem is locally solved (I think, the error message
is gone and hello-world.scm works) by using --no-grafts in every Guix
invocation.  What I could/will still do is collect gdk-pixbuf debug info
in the same directory as locally patched gtk debug info.

>> Setting the GTK_A11Y env var does not make a difference (except the
>> warning is gone) and on my x86_64 GNOME setup there never is a warning
>> (although GTK_A11Y is unset, but d-bus is there).
>
> I never thought in any moment that this warning was in any way 'linked'
> to the problem, i did say and re-state here, that guix should be fixed
> so it does not complain it can't find d-bus - terrible imo, it can lead
> to other problems - till then, locally define the env var so we get
> this annoying warning out of the way

Well the d-bus service cannot be a requirement of a Guix package.  It
could be a requirement of a Guix service, but g-golf is not a service
but just a package.

Regards,
Florian