save_image() shouldn't throw exceptions because if the 'then_die'
argument is t it would leave factor in an inconsistent state. So
therefore move_file() should be fixed and raw_fclose() called instead of
safe_fclose().
* Clear faulting_p from a safepoint rather than inside general_error, because jumping into unwind-native-frames could blow up.
* Handle multiple faults from fatal_error by breakpointing. Is there anything else we can safely do at that point?
* Verify memory protection faults in the top half of the signal handlers because signal dispatch could fault. Treat memory faults during gc or fep as fatal errors.
* Add a function factor_vm::abort() that restores the default SIGABRT handler and ::abort()s. Use it from fatal_error() so we get useful context from gdb and so the user gets feedback from the system crash reporter that Factor blew up and didn't just disappear.
* In factorbug(), don't proceed with .s .r .c if it would be unsafe to do so.
* Don't pile on signals if we've already called fatal_error().
The stdin_loop thread will keep trying to consume input unless we stop it by sending it a signal. Use SIGUSR2 to stop the read syscall and a mutex to hold up the loop while the fep is active.
- Move forward declarations of 'struct factor_vm' to one place
- Rename template parameters from T and TYPE to descriptive names. New convention: CamelCase for template parameters
- Change some higher-order functions taking function pointers into templates, and define classes overriding operator(). There's a bit of new boilerplate here but its more consistent than the old mish-mash approaches
- Put GC state into a gc_state struct
- Use exceptions instead of longjmp for non-local control transfer in GC
- In code GC, instead of interleaving code block tracing with copying, add code blocks which need to be revisited to an std::set stored in the gc_state
Björn Lindqvist
Doug Coleman
Erik Charlebois
Joe Groff
Slava Pestov
Slava Pestov
Slava Pestov
Phil Dawes
Philipp Brüschweiler
Slava Pestov