p5.16-wx: segfaults when running command

[mf2k (Frank Schima)]

p5.16-wx segfaults when the following command is issued:

perl -e 'use Wx qw(:everything); use Wx::Html;'

Cc'ing mojca for possible comment.

[mojca (Mojca Miklavec)]

Sorry, no clue.

Process:         perl5.16 [92649]
Path:            /opt/local/bin/perl5.16
Identifier:      perl5.16
Version:         ??? (???)
Code Type:       X86-64 (Native)
Parent Process:  bash [92434]

Date/Time:       2014-01-29 01:27:38.413 +0100
OS Version:      Mac OS X 10.7.5
Report Version:  9

Crashed Thread:  0  Dispatch queue: com.apple.main-thread

Exception Type:  EXC_BAD_ACCESS (SIGSEGV)
Exception Codes: KERN_INVALID_ADDRESS at 0x0000000000000000

VM Regions Near 0:
--> 
    __TEXT                 00000001089b8000-00000001089b9000 [    4K] r-x/rwx SM=COW  /opt/local/bin/perl5.16

Application Specific Information:
objc[92649]: garbage collection is OFF

Thread 0 Crashed:: Dispatch queue: com.apple.main-thread
0   libwx_baseu-2.9.dylib         	0x0000000109860a0f wxPluginLibrary::UpdateClasses() + 43
1   libwx_baseu-2.9.dylib         	0x0000000109860993 wxPluginLibrary::wxPluginLibrary(wxString const&, int) + 177
2   libwx_baseu-2.9.dylib         	0x000000010986147a wxPluginManager::LoadLibrary(wxString const&, int) + 654
3   Wx.bundle                     	0x0000000108d11efe _ZL18XS_Wx__load_pluginP11interpreterP2cv + 430
4   libperl.dylib                 	0x0000000108a48dbc Perl_pp_entersub + 2332
5   libperl.dylib                 	0x0000000108a420e6 Perl_runops_standard + 22
6   libperl.dylib                 	0x00000001089da7ed Perl_call_sv + 733
7   libperl.dylib                 	0x00000001089d7043 Perl_call_list + 547
8   libperl.dylib                 	0x00000001089ccac3 S_process_special_blocks + 211
9   libperl.dylib                 	0x00000001089cc22a Perl_newATTRSUB_flags + 3690
10  libperl.dylib                 	0x00000001089c7639 Perl_utilize + 1193
11  libperl.dylib                 	0x0000000108a02736 Perl_yyparse + 934
12  libperl.dylib                 	0x00000001089d959f perl_parse + 8767
13  perl5.16                      	0x00000001089b8d6e main + 158
14  perl5.16                      	0x00000001089b8cc4 start + 52

Thread 1:: Dispatch queue: com.apple.libdispatch-manager
0   libsystem_kernel.dylib        	0x00007fff8b5e27e6 kevent + 10
1   libdispatch.dylib             	0x00007fff8c329786 _dispatch_mgr_invoke + 923
2   libdispatch.dylib             	0x00007fff8c328316 _dispatch_mgr_thread + 54

Thread 2:
0   libsystem_kernel.dylib        	0x00007fff8b5e2192 __workq_kernreturn + 10
1   libsystem_c.dylib             	0x00007fff89670594 _pthread_wqthread + 758
2   libsystem_c.dylib             	0x00007fff89671b85 start_wqthread + 13

Thread 3:
0   libsystem_kernel.dylib        	0x00007fff8b5e2192 __workq_kernreturn + 10
1   libsystem_c.dylib             	0x00007fff89670594 _pthread_wqthread + 758
2   libsystem_c.dylib             	0x00007fff89671b85 start_wqthread + 13

Thread 0 crashed with X86 Thread State (64-bit):
  rax: 0x00007fd3da475808  rbx: 0x00007fd3d8e09e70  rcx: 0x0000000000000001  rdx: 0x00007fff685b6ee0
  rdi: 0x00007fd3da4757f0  rsi: 0x0000000000000301  rbp: 0x00007fff685b6ff0  rsp: 0x00007fff685b6fa0
   r8: 0x0000000000000014   r9: 0x0000000000000efc  r10: 0x0000000000000000  r11: 0x00007fd3da475808
  r12: 0x00007fff685b6fc0  r13: 0x0000000108c77d60  r14: 0x00007fff685b6fa0  r15: 0x0000000000000000
  rip: 0x0000000109860a0f  rfl: 0x0000000000010287  cr2: 0x0000000000000000
Logical CPU: 0

I would say "ask on the upstream mailing list", but I'm aware that nobody's answering. You could try on the wxWidgets side, but I'm also not sure if they will bother to answer any questions related to version 2.9.4.

[mf2k (Frank Schima)]

Thanks for looking at this. Hopefully an update upstream that fixes this will eventually happen.

[mojca (Mojca Miklavec)]

As noted in #42073 there is a high chance that the latest version of wxPerl committed in r118413 fixed the problem. But please test on your own machine and report if there are still problems (or close the ticket).

[mf2k (Frank Schima)]

Yes, it appears to no longer segfault. Thanks.