https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=273291
--- Comment #18 from Tatsuki Makino <tatsuki_mak...@hotmail.com> ---
(In reply to Tomoaki AOKI from comment #13)
Those are just suggestions to see if there are any problems when using GPUs.
I don't know how to use --headless either :)
However, it seems to us that the problem is the part about the GUI.
Then, my explanation about lldb in ML was not enough.
Type c into lldb only if the program stops at an irrelevant place, if the
program stops with a crash, it should be typed bt all immediately.
For example, the following
> lldb -o r -- /tmp/a.out
(lldb) target create "/tmp/a.out"
Current executable set to '/tmp/a.out' (x86_64).
(lldb) r
Process 6041 stopped
* thread #1, name = 'a.out', stop reason = signal SIGSEGV: invalid address
(fault address: 0x0)
frame #0: 0x00000000002018dd a.out`main at test.c:4:11
1 int
2 main()
3 {
-> 4 int ga = *((int *)0);
5 return 0;
6 }
Process 6041 launched: '/tmp/a.out' (x86_64)
(lldb) bt all
* thread #1, name = 'a.out', stop reason = signal SIGSEGV: invalid address
(fault address: 0x0)
* frame #0: 0x00000000002018dd a.out`main at test.c:4:11
frame #1: 0x00000000002016d0 a.out`_start + 256
When c is typed here, the program runs to the point of dumping core and exits.
(lldb) c
Process 6041 resuming
Process 6041 exited with status = 11 (0x0000000b)
(lldb) q
If q is typed instead of c, the program is in the middle of execution and will
be forced to terminate.
(lldb) q
Quitting LLDB will kill one or more processes. Do you really want to proceed:
[Y/n]
I think the backtrace we get after reloading the core is still a little out of
place.
So, sorry, but would you please collect some more backtraces?
--
You are receiving this mail because:
You are the assignee for the bug.
