Re: Compiler crashes when taking address of function pointer

[Ori Bernstein <ori@xxxxxxxxxxxxxx>]

Ok, for now:

	var p : byte#
	var i : std.intptr

	p = somefunc castto(byte#)
	i = somefunc castto(std.intptr)

Should work error free. Figuring out what exactly function addresses mean
semantically, and how they interact with closures, is something I'll have
to think about a bit more.
	

On Fri, 16 Jan 2015 22:55:48 -0800
Ori Bernstein <ori@xxxxxxxxxxxxxx> wrote:

> Sure; for a simple fix, I can allow casting from functions to pointers
> and back. I'm just trying to sort out what it would mean to, eg, deref
> a function pointer.
> 
> And what happens to that code if/when I figure out environment capture
> for closures.
> 
> On Fri, 16 Jan 2015 23:38:04 -0600
> Ryan <rymg19@xxxxxxxxx> wrote:
> 
> > 
> > 
> > Ori Bernstein <ori@xxxxxxxxxxxxxx> wrote:
> > >Interesting. That shouldn't happen. It's crashing because the backend
> > >puts
> > >function addresses into registers as rvalues if they're not being
> > >called.
> > >
> > >I haven't hit it because I haven't needed pointer to a function yet;
> > >functions already can be passed around without taking their address:
> > >
> > >	const each = {sl : int[:], f : (int -> void)
> > >		...
> > >	}
> > >
> > >	each([1,2,3][:], somefunc)
> > 
> > Well, it has to do with the kernel writing. I need to split a function's address into the higher and lower bits.
> > 
> > I think there's a way I can get around it, though. Just though I should report this anyway.
> > 
> > >
> > >should work just fine.
> > >
> > >On Fri, 16 Jan 2015 14:52:29 -0600
> > >Ryan Gonzalez <rymg19@xxxxxxxxx> wrote:
> > >
> > >> Example:
> > >> 
> > >> const g = {; ->}
> > >> const f = {; var addr = &g}
> > >> 
> > >> I get this:
> > >> 
> > >> ryan@DevPC-LX:~/langtest/myr$ 6m bug.myr
> > >> 6m: gengas.c:128: locprint: Assertion `(spec == 'r' &&
> > >isintmode(l->mode))
> > >> || (spec == 'f' && isfloatmode(l->mode)) || spec == 'v' || spec ==
> > >'x' ||
> > >> spec == 'u'' failed.
> > >> Aborted (core dumped)
> > >> ryan@DevPC-LX:~/langtest/myr$
> > >> 
> > >> -- 
> > >> Ryan
> > >> If anybody ever asks me why I prefer C++ to C, my answer will be
> > >simple:
> > >> "It's becauseslejfp23(@#Q*(E*EIdc-SEGFAULT. Wait, I don't think that
> > >was
> > >> nul-terminated."
> > >> Personal reality distortion fields are immune to contradictory
> > >evidence. -
> > >> srean
> > >> Check out my website: http://kirbyfan64.github.io/
> > 
> > -- 
> > Sent from my Android phone with K-9 Mail. Please excuse my brevity.
> > Check out my website: http://kirbyfan64.github.io/
> > 
> 
> 
> -- 
> Ori Bernstein <ori@xxxxxxxxxxxxxx>
> 


-- 
Ori Bernstein <ori@xxxxxxxxxxxxxx>