Date: Sun, 22 Oct 2006 21:36:08 -0400 From: Joe Marcus Clarke <marcus@marcuscom.com> To: Jeremy Messenger <mezz7@cox.net> Cc: freebsd-stable@freebsd.org Subject: Re: Do anyone has any problem with sem_open() crash? Message-ID: <1161567368.30822.31.camel@shumai.marcuscom.com> In-Reply-To: <op.thun54ec9aq2h7@mezz.mezzweb.com> References: <op.thun54ec9aq2h7@mezz.mezzweb.com>
next in thread | previous in thread | raw e-mail | index | archive | help
[-- Attachment #1 --] On Sun, 2006-10-22 at 20:09 -0500, Jeremy Messenger wrote: > Hello folks, > > Last night, I have upgraded from July RELENG_6 to last night. I have > reinstalled all of ports and I noticed that the log/messages is collecting > a few of 'pid 14699 (conftest), uid 0: exited on signal 12 (core dumped)'. > I dig in the configure and found a location of crash. It is from > sem_open(). I don't get that before I upgraded the RELENG_6. > > http://people.freebsd.org/~mezz/test-sem_open.cc (took from configure's > conftest) > > =============================== > % c++ -o test-sem_open -O2 -fno-strict-aliasing -pipe -g -pthread -pipe > test-sem_open.cc > % ./test-sem_open > [1] 5761 invalid system call (core dumped) ./test-sem_open > > % grep sem_open /var/log/messages > Oct 22 19:44:35 mezz kernel: pid 5761 (test-sem_open), uid 1001: exited on > signal 12 (core dumped) > Oct 22 20:00:45 mezz kernel: pid 6140 (test-sem_open), uid 1001: exited on > signal 12 (core dumped) > =============================== > > I have decided to login in a jail that has RELENG_6 from July in the same > local machine that have most debug enable. > > =============================== > $ gdb test-sem_open test-sem_open.core > > Core was generated by `test-sem_open'. > Program terminated with signal 12, Bad system call. > > #0 0x281bf297 in ksem_open () at ksem_open.S:2 > 2 ksem_open.S: No such file or directory. > in ksem_open.S > [New LWP 100120] > > > (gdb) bt > #0 0x281bf297 in ksem_open () at ksem_open.S:2 > #1 0x281b5cd8 in __sem_open (name=0x804868e "ace_semaphore_foo", > oflag=2560) > at /usr/src/lib/libc/gen/sem.c:217 > #2 0x080485ee in main () at test-sem_open.cc:168 > Current language: auto; currently asm > > > (gdb) bt full > #0 0x281bf297 in ksem_open () at ksem_open.S:2 > No locals. > #1 0x281b5cd8 in __sem_open (name=0x804868e "ace_semaphore_foo", > oflag=2560) > at /usr/src/lib/libc/gen/sem.c:217 > sem = (sem_t *) 0xbfbfec9c > s = 0x4e > semid = 1 > mode = 78 > value = 1 > #2 0x080485ee in main () at test-sem_open.cc:168 > s = (sem_t *) 0xa00 > =============================== > > BTW: Please add me in CC, I am not on the list. You need to make sure you have "options P1003_1B_SEMAPHORES" in your kernel. These POSIX IPC constructs are labeled as being very experimental, and this is not a default option. Joe -- PGP Key : http://www.marcuscom.com/pgp.asc [-- Attachment #2 --] -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (FreeBSD) iD8DBQBFPByIb2iPiv4Uz4cRApTiAKCKbvzR5G6fgil7+O8B1pUrNhs05wCfSfak 1/xdho9/rIMc9cghSj2sWGk= =1Twd -----END PGP SIGNATURE-----
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1161567368.30822.31.camel>