Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-08 Thread John Reiser
On 6/8/25 1:34 AM, Martin Gansser wrote: I always reproduce the error as follows: gdb speed-dreams-2 then select Race - Championchip select player, then select start and start again. Please be specific. There is one player already listed in light orange in a box of players. Or, clicking on the

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-08 Thread Martin Gansser
I always reproduce the error as follows: gdb speed-dreams-2 then select Race - Championchip select player, then select start and start again. Then press return to start the game. Then end the game with the ESCAPE key or by closing the game window with a mouse click. -- __

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-07 Thread John Reiser
What is the recipe to reproduce the crash? I see no crash running speeddreams-2.4.2 from rawhide on a fresh Fedora 42: click anywhere in the window with 3 cars, Quit, confirm. Also works with valgrind. It seems to me that many of the "???" return addresses probably are on pages that have been

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-06 Thread Martin Gansser
have attached all the output of the commands $ strace -f -o strace.out -e trace=memory speed-dreams-2 [1] $ valgrind --track-origins=yes --freelist-vol=$BIG_NUMBER --log-file=SD-valgrin.log speed-dreams-2 [2/3] [1] https://martinkg.fedorapeople.org/ErrorReports/strace.out [2] https://martinkg.fe

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-05 Thread John Reiser
   strace -f -o strace.out -e trace=memory speed-dreams-2 Memcheck (valgrind) may help if its queue of blocks that have been released (free(), munmap()), but not yet made available for re-allocation by memcheck, is long enough. Try: valgrind -track-origins=yes --freelist-vol=$BIG_NUMBER sp

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-05 Thread John Reiser
On 6/5/25 12:00 AM, Martin Gansser wrote: [1] https://martinkg.fedorapeople.org/ErrorReports/gdb.txt Thread 1 (Thread 0x769b9fc0 (LWP 201755) "speed-dreams-2"): #0 0x7fffad1687dc in ??? () #1 0x7fffdb60 in ??? () #2 0x77a90592 in _int_free (av=, p=, have_lock=0) at

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-05 Thread Martin Gansser
Enclosed is the requested information [1] [1] https://martinkg.fedorapeople.org/ErrorReports/gdb.txt Regards Martin -- ___ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-le...@lists.fedoraproject.org Fedora Co

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-04 Thread John Reiser
On 6/4/25 1:21 AM, Martin Gansser wrote: I have now compiled locally again with only the following option: <> $ export DEBUGINFOD_URLS="https://debuginfod.fedoraproject.org/"; $ gdb /usr/bin/speed-dreams-2 backtrace: [1] https://martinkg.fedorapeople.org/ErrorReports/gdb.txt Thread 1 "sp

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-04 Thread Dan Horák
On Wed, 04 Jun 2025 06:59:07 - "Martin Gansser" wrote: > have set %global _lto_cflags %{nil} as you suggested. > [1] https://martinkg.fedorapeople.org/ErrorReports/gdb.txt sounds like the installed glibc debuginfo doesn't match the installed runtime glibc rpm. Be aware that debuginfos are us

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-04 Thread Martin Gansser
I have now compiled locally again with only the following option: %build %cmake -DCMAKE_BUILD_TYPE:STRING=RelWithDebInfo \ -DCMAKE_SKIP_RPATH:BOOL=OFF\ -DOPTION_DEBUG:STRING=ON \ -DCMAKE_SHARED_LINKER_FLAGS="-Wl,--as-needed" \

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-03 Thread Martin Gansser
have set %global _lto_cflags %{nil} as you suggested. [1] https://martinkg.fedorapeople.org/ErrorReports/gdb.txt Regards Martin -- ___ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-le...@lists.fedoraproject.or

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-03 Thread Vitaly Zaitsev via devel
On 02/06/2025 12:09, Martin Gansser wrote: I have installed speed-dreams-2.4.2 and noticed that a SIGSEGV, segmentation fault Can you try building without LTO optimizations? %global _lto_cflags %{nil} -- Sincerely, Vitaly Zaitsev (vit...@easycoding.org) -- _

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-03 Thread Vitaly Zaitsev via devel
On 03/06/2025 11:13, Florian Weimer wrote: If you use RelWithDebInfo instead, as Martin suggested, you'll get a better debugging experience. On Fedora, you won't notice any difference between the Release and RelWithDebInfo build configurations, since Fedora's build flags already include the -

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-03 Thread Martin Gansser
the build section of the speed-dreams.spec file looks like this on my local build: "%build %cmake -DCMAKE_BUILD_TYPE:STRING=RelWithDebInfo \ -DCMAKE_SKIP_RPATH:BOOL=OFF\ -DOPTION_DEBUG:STRING=ON \ -DCMAKE_SHARED_LINKER_FLAGS="-W

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-03 Thread Miroslav Suchý
Dne 02. 06. 25 v 1:40 odp. Martin Gansser napsal(a): speed-dreams-debuginfo-2.4.2-1.fc42.x86_64 is already installed. Regards Martin No need to install it locally with debuginfod. See https://fedoraproject.org/wiki/Debuginfod and https://developers.redhat.com/blog/2019/10/14/introducing-debu

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-03 Thread Florian Weimer
* Martin Gansser: > i think the current backtrace log [1] is also useless. > [1] ]https://martinkg.fedorapeople.org/ErrorReports/gdb2.txt > > Yes, i used on the koji buidl this Option: > '-DCMAKE_BUILD_TYPE:STRING=Release' If you use RelWithDebInfo instead, as Martin suggested, you'll get a bet

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-03 Thread Martin Gansser
maybe the next steps helps, i am not a debugging specialist. Thread 1 "speed-dreams-2" received signal SIGSEGV, Segmentation fault. 0x7fffb11687dc in ?? () (gdb) bt full #0 0x7fffb11687dc in ??? () #1 0x7fffdb60 in ??? () #2 0x77a90592 in _int_free (av=, p=, have_lock=

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-02 Thread Martin Gansser
i think the current backtrace log [1] is also useless. [1] ]https://martinkg.fedorapeople.org/ErrorReports/gdb2.txt Yes, i used on the koji buidl this Option: '-DCMAKE_BUILD_TYPE:STRING=Release' -- ___ devel mailing list -- devel@lists.fedoraproject.or

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-02 Thread Leigh Scott
Those options look ok, I note your koji build used '-DCMAKE_BUILD_TYPE:STRING=Release' -- ___ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-le...@lists.fedoraproject.org Fedora Code of Conduct: https://docs.f

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-02 Thread Martin Gansser
is this correct or is there something missing to the options: %build %cmake -DCMAKE_BUILD_TYPE:STRING=RelWithDebInfo \ -DCMAKE_SKIP_RPATH:BOOL=OFF\ -DOPTION_DEBUG:STRING=ON \ -DCMAKE_SHARED_LINKER_FLAGS="-Wl,--as-needed" \

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-02 Thread Leigh Scott
Check your build flags to make sure '-g' is included and cmake doesn't strip during build. -- ___ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-le...@lists.fedoraproject.org Fedora Code of Conduct: https://d

speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-02 Thread Martin Gansser
Hi, I have installed speed-dreams-2.4.2 and noticed that a SIGSEGV, segmentation fault [1]. occurs when exiting the program. Unfortunately, the thread in question (speed-dreams-2) does not display any function names, which is probably due to missing symbols: in the upstream ticket https://for

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-02 Thread Martin Gansser
speed-dreams-debuginfo-2.4.2-1.fc42.x86_64 is already installed. Regards Martin -- ___ devel mailing list -- devel@lists.fedoraproject.org To unsubscribe send an email to devel-le...@lists.fedoraproject.org Fedora Code of Conduct: https://docs.fedorap

Re: speed-dreams-2.4.2: How to determine the cause of a SIGSEGV signal

2025-06-02 Thread Tomasz Torcz
On Mon, Jun 02, 2025 at 10:09:43AM -, Martin Gansser wrote: > Hi, > > I have installed speed-dreams-2.4.2 and noticed that a SIGSEGV, segmentation > fault [1]. occurs when exiting the program. > Unfortunately, the thread in question (speed-dreams-2) does not display any > function names, whi