www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.bugs - void.init -> internal error

reply Benjamin Herr <ben 0x539.de> writes:
$ cat void.d
private import std.stdio;
void f() { writef(void.init); }

$ dmd void.d
Internal error: ../ztc/cod1.c 2498

dmd 0.101, linux 32bit

I was just trying to find out what interesting things one could do with voids
after following the how-to-slice-structs-to-void[] thread.

Otherwise, looking forward to a D compiler that builds on amd64 :)
Sep 10 2004
parent reply Sean Kelly <sean f4.ca> writes:
In article <chtl8a$2509$1 digitaldaemon.com>, Benjamin Herr says...

Otherwise, looking forward to a D compiler that builds on amd64 :)


I'd been meaning to ask about this.  What exactly doesn't work on Amd64?  And is
there any hope of this being fixed by 1.0?


Sean
Sep 10 2004
parent Benjamin Herr <ben 0x539.de> writes:
In article <chtm7r$25g4$1 digitaldaemon.com>, Sean Kelly says...

In article <chtl8a$2509$1 digitaldaemon.com>, Benjamin Herr says...

Otherwise, looking forward to a D compiler that builds on amd64 :)


I'd been meaning to ask about this.  What exactly doesn't work on Amd64?  And is
there any hope of this being fixed by 1.0?


The dmd distribution was compiled for x86, and I would need an x86-64 binary.
Without it being free software, I cannot even compile it enough to see what does
not work. :)
I do not think it is much work. Most programs I come across are written in a
non-platform dependant way, so I can use most applications like gcc, Firefox,
ruby etc without problems. It would probably just require a recompile with an
amd64 compiler and perhaps fixing minor bugs. 
Considering that D development appears to be pretty Win32 centered, I guess
portability is not a top priority, so I think Walter has more important things
to do.

If the gdc folks want me to do some bug reporting, I could do that, though;
although I would not be sure whether problems are caused by my failure to
properly compile gcc
Sep 11 2004