Perl 6 - the future is here, just unevenly distributed

IRC log for #divine, 2015-12-06

| Channels | #divine index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
13:37 zbeasnyy joined #divine
23:02 xtomast1 joined #divine
23:03 xtomast1 tak to s tou globalni promennou a jejim nastavovanim pomoci __attribute__((constructor)) zda se funguje, ale v souvislosti se mi objevil takovy podivny problem
23:03 xtomast1 _initialized_ = new (nofail) bool[_mpi_num_processes] {};
23:03 xtomast1 _finalized_ = new (nofail) bool[_mpi_num_processes] {};
23:03 xtomast1 _buffer_attached_ = new (nofail) bool[_mpi_num_processes] {};
23:03 xtomast1 _attached_buffer_ = new (nofail) char*[_mpi_num_processes];
23:03 xtomast1 v tomto kodu dostanu na poslednim radku chybu
23:03 xtomast1 UNDEFINED VALUE (thread 0): <_divine_mpi_global_initialize()> [ mpi.cpp:6329 ] << %62 = select i1 %60, i64 -1, i64 %61, !dbg !110198 >>
23:04 xtomast1 ...a netusim co s tim. prijde mi to podivne proto, ze jednak u tech predchozich radku je podle vseho vseho vsechno v poradku, a jednak kdyz tam misto _mpi_num_processes dam primo "2" (a pred to assert(_mpi_num_processes == 2)), tak to projde
23:26 mornfall joined #divine
23:35 xstill víš že bez těch záverek se neinicializuje ten objekt na nuly jo? Nevím jestli to může nějak ovlivnit…
23:35 xstill jinak bych asi potřeboval vidět víc kódu a co divine říká…

| Channels | #divine index | Today | | Search | Google Search | Plain-Text | summary