[Octave-bug-tracker] [bug #34663] OpenBSD Build Failure with regards to sigaddset()

[Edd Barrett]

Follow-up Comment #5, bug #34663 (project octave):

Hi John,

Having worked around #34725, we should figure out what is wrong here. I had
previously applied a patch which worked around a build error. I am attaching
it.

With this patch applied, I get the following seg:

(gdb) run
Starting program: /usr/ports/pobj/octave-3.4.3/octave-3.4.3/src/.libs/octave 

Program received signal SIGSEGV, Segmentation fault.
[Switching to process 1419, thread 0x88dd3400]
0x0948d51d in octave_set_signal_handler (sig=2, 
    handler=0x948e050 <sigint_handler>, restart_syscalls=true)
    at sighandlers.cc:225
225       gnulib::sigemptyset (&act.sa_mask);
Current language:  auto; currently c++
(gdb) bt
#0  0x0948d51d in octave_set_signal_handler (sig=2, 
    handler=0x948e050 <sigint_handler>, restart_syscalls=true)
    at sighandlers.cc:225
#1  0x0948d600 in octave_catch_interrupts () at sighandlers.cc:460
#2  0x0948d6c8 in install_signal_handlers () at sighandlers.cc:529
#3  0x0945cce5 in octave_main (argc=1, argv=0xcfbd89e0, embedded=0)
    at octave.cc:665
#4  0x1c00093a in main (argc=536914932, argv=0x88fdb078) at main.c:35


What is supposed to happen here, am I supposed to be using gnulib
replacements? OpenBSD has implementations of both sigaddset and sigemptyset.

It looks here like I am mixing implementations, so in a way it does not
surprise me that this does not work.

Can you comment? Did the build system do the right thing in choosing gnulib?
If so, the original assembler message is our blocker, otherwise, the build
system erroneously chose to replace system functions.

Can you comment?

(file #24301)
    _______________________________________________________

Additional Item Attachment:

File name: patch-src_sighandlers_cc       Size:0 KB


    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?34663>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/