alan buckley wrote:
I've got an unexpected crash when throwing an
exception. I've created a test progam below. I
believe it's a problem in GCC4 as a program
compiled with GCC3 has similar code that works
as expected.
I had some debug turned on in my shared UnixLib, and
built a dynamic binary, so have a little more information,
below. I don't know why there isn't a meaningful
backtrace on the static binary.
in canon
pre call
Throwing error Filing system ADFS: must be given a filename
"/usr/src/gccsdk/gcc4/srcdir/libunixlib/gcccompat/galloca.c", line 190:
__gcc_alloca_restore: Assertion failed: chunk != NULL
-- __ul_raise_signal: deliver pending signals
-- post_signal: Deliver pending signals. Pending 00000000, blocked 00000000
-- post_signal: pending signals delivered
-- __ul_raise_signal: returning
-- __ul_raise_signal: sigtate 00000000, signo 6 (Aborted)
-- post_signal: sigstate=000160c8, signo 6 (Aborted)
-- post_signal: handler=00000000 (SIG_DFL)
-- post_signal: action taken = core
-- post_signal: term/core
Fatal signal received: Aborted
Stack backtrace:
Running thread 0x16494
( a6e68) pc: 5c90d8d4 lr: 5c90e014 sp: a6e6c __write_backtrace()
( a6ee0) pc: 5c90dad8 lr: 5c90c2cc sp: a6ee4
__unixlib_raise_signal()
( a6ef4) pc: 5c90c298 lr: 5c886d14 sp: a6ef8 raise()
( a6f08) pc: 5c886cc0 lr: 5c9060cc sp: a6f0c abort()
( a6f28) pc: 5c906038 lr: 5c93a548 sp: a6f2c __assert2()
( a6f50) pc: 5c93a418 lr: 8ee8 sp: a6f54 __gcc_alloca_restore()
( a6fa8) pc: 8d9c lr: 8f68 sp: a6fac
canonicalise(std::string)()
( a6ff4) pc: 8f28 lr: 5c95c89c sp: a6ff8 main()
_______________________________________________
GCCSDK mailing list [email protected]
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK
