Re: Preventing stack overflows with alloca.

[Kenichi Handa]

In article <address@hidden>, address@hidden (Kim F. Storm) writes:

> Since old_value is not a lisp object here,
> we would need to fix (hack) GC like this:

>   for (bind = specpdl; bind != specpdl_ptr; bind++)
>     {
>       mark_object (bind->symbol);
>       if (bind->func != safe_alloca_unwind)
>         mark_object (bind->old_value);
>     }

I now remembered why I didn't work on it further.  That's
because I didn't find the above simple solution!  So, I was
going to change `unused' of struct specbinding to void *val,
make record_unwind_protect_c which accepts void * instead of
Lisp_Object as ARG, make unbind_to to call func with val
instead of old_value of val is not NULL, etc... and noticed
that it's a too big change for the current rather simple
problem.  :-(

> A sample use would look like this:

> Lisp_Object
> string_make_multibyte (string)
>      Lisp_Object string;
> {
>   unsigned char *buf;
>   int nbytes;
>   Lisp_Object ret;
>   USE_SAFE_ALLOCA;

>   if (STRING_MULTIBYTE (string))
>     return string;

>   nbytes = count_size_as_multibyte (SDATA (string),
>                                   SCHARS (string));
>   /* If all the chars are ASCII, they won't need any more bytes
>      once converted.  In that case, we can return STRING itself.  */
>   if (nbytes == SBYTES (string))
>     return string;

>   SAFE_ALLOCA (buf, nbytes);

>   copy_text (SDATA (string), buf, SBYTES (string),
>            0, 1);

>   ret = make_multibyte_string (buf, SCHARS (string), nbytes);

>   SAFE_FREE (nbytes);

>   return ret;
> }

I like it.

---
Ken'ichi HANDA
address@hidden