Thomas Zellerin wrote:
>My Linux can, if I give the handler some stack to use and tell him to
>use it. Is this what you need?
Thanks for pointing me to man sigaltstack(). The manpage says:
reaches a limit established by a call to setr=AD
limit(RLIMIT_STACK, &rlim). If the standard stack is
exhausted, the kernel sends the process a SIGSEGV signal.
In these circumstances the only way to catch this signal
is on an alternate signal stack.
Apparently, that's how GDB works to catch the signal while executing
p zout *STACK
If it's so easy, then I'm really wondering why libsigsegv does not =
provide exactly that functionality. Build problem? (see below)
This makes me run ./configure on the libsigsegv source tree on a Suse =
Linux 8.0 i386 box:
checking if the system supports catching stack overflow... yes -- !?!
make produces a sigsegv.h which says:
# define HAVE_STACK_OVERFLOW_RECOVERY 1
All 4 tests passed
Grr. So why does the Ubuntu/Debian libsigsegv package *not* define =
I think I've installed Peter's libsigsegv-2.2 package (can't check =
Peter? Build problem?
I'll build a libsigsegv on the i386 Ubuntu box asap and report results.
PS: how comes I receive that e-mail without the [clisp-devel] subject =