Re: Inlining policy

[Stefan Monnier]

> Actually I think moving the lisp_h_* into the functions could also make
> these easier to debug in case.  I find a little odd to optimize for
> speed -O0, in the end is the exact reason why -Og exists.  But I
> understand there's a trade-off here.

FWIW, I use `-Og` nowadays and can't remember the last time I used
`-O0`, so I'd be happy with a solution that doesn't work so well for
`-O0` as long as it works well for `-Og`.


        Stefan