Re: [CRASH] 21.2-b5 with Jonathan's gc crash fix applied
Norbert Koch <n.koch(a)delta-ii.de> writes:
> I think the idea is to invoke the function immediately *before*
> the crash, not after it. If the crash is happening due to an
> assert or something like that, under Unix I'd put a breakpoint on
> assert_failed() and invoked debug_backtrace() from there.
Dunno, perhaps a good idea would be to call the function before and
after the crash (perchance, there is an after :-).
On Unix, you cannot do that. When the process SIGSEGV's and dies of
it, you cannot call functions from the debugger (you can still examine
the variable and stuff, but no function-calling.)