Date: Mon, 5 Oct 2020 17:20:55 +0000 From: Dave Baukus <daveb@spectralogic.com> To: "freebsd-threads@freebsd.org" <freebsd-threads@freebsd.org> Subject: SIGSEV in thr_malloc_lock() via __Tthr_mutex_init() Message-ID: <f09a55c8-c0ba-52bd-eb64-cfd9c1ff52c9@spectralogic.com>
next in thread | raw e-mail | index | archive | help
On FreeBSD stable12, the testparm utility of a local build of samba413 accesses an invalid curthread as returned from _get_curthread(). Breakpoint 2, thr_malloc_lock (curthread=0x802c057d0) at lib/libthr/thread/thr_malloc.c:64 64 if (curthread == NULL) (gdb) bt #0 thr_malloc_lock (curthread=0x802c057d0) at lib/libthr/thread/thr_malloc.c:66 #1 __thr_calloc (num=1, size=96) at lib/libthr/thread/thr_malloc.c:88 #2 0x00000008014330a3 in mutex_init (mutex=0x80106f008, mutex_attr=<optimized out>, calloc_cb=<optimized out>) at lib/libthr/thread/thr_mutex.c:295 #3 __Tthr_mutex_init (mutex=0x80106f008, mutex_attr=<optimized out>) at lib/libthr/thread/thr_mutex.c:395 #4 0x000000080203ef2c in gnutls_system_mutex_init () from /usr/local/lib/libgnutls.so.30 #5 0x0000000802038773 in _gnutls_global_init () from /usr/local/lib/libgnutls.so.30 #6 0x0000000802038b37 in lib_init () from /usr/local/lib/libgnutls.so.30 #7 0x000000080103564d in ?? () from /libexec/ld-elf.so.1 #8 0x0000000801050b00 in ?? () thr_malloc_lock() will core when it attempts curthread->locklevel++ because curthread (0x802c057d0) is in a RO section of libnettle: (gdb) info proc mappings process 40081 Mapped address spaces: Start Addr End Addr Size Offset Flags File ... 8< snip 0x802c03000 0x802c1f000 0x1c000 0x0 r-- CN-- /usr/local/lib/libnettle.so.8.0 0x802c1f000 0x802c40000 0x21000 0x1c000 r-x CN-- /usr/local/lib/libnettle.so.8.0 Anyone have a clue as to what's going on or a suggestion on what may be wrong with my build ? Thanks -- Dave Baukus
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?f09a55c8-c0ba-52bd-eb64-cfd9c1ff52c9>