Can anyone tell me what the situation is with 64-bit support in DMD? Like when is it likely to happen?

Graham St Jack wrote:
> Can anyone tell me what the situation is with 64-bit support in DMD? Like when is it likely to happen?

I  don't think it's likely to happen, because DMD uses a backend that already existed before D and that has only 32bit support. But there's always GDC with 64bit support, at least on windows.

doob wrote:
> Graham St Jack wrote:
>> Can anyone tell me what the situation is with 64-bit support in DMD? Like when is it likely to happen?
> 
> I  don't think it's likely to happen, because DMD uses a backend that already existed before D and that has only 32bit support. But there's always GDC with 64bit support, at least on windows.

I've got hopes that one day an LLVM-based D compiler will be the one D compiler to rule them all.
(There's a start of one limping along at http://www.dsource.org/projects/llvmdc).

--bb

> I've got hopes that one day an LLVM-based D compiler will be the one D compiler to rule them all.
> (There's a start of one limping along at http://www.dsource.org/projects/llvmdc).
> 
> --bb

Very good news!  I hope this gets more attention.

> I've got hopes that one day an LLVM-based D compiler will be the one D
> compiler to rule them all.
> (There's a start of one limping along at
> http://www.dsource.org/projects/llvmdc).

Sounds great - looking forward to it.

Graham St Jack wrote:
>> I've got hopes that one day an LLVM-based D compiler will be the one D
>> compiler to rule them all.
>> (There's a start of one limping along at
>> http://www.dsource.org/projects/llvmdc).
> 
> Sounds great - looking forward to it.

Yeh, me too.  But I'm afraid that unless it becomes someone's full-time obsession, realistically it's never going to be ready for prime time.

Thomas L. Olsen has gotten it off to a good start as his hobby project, despite apparently starting from zero knowledge of how LLVM works.  But I don't think his spare time effort is really going to be enough to create a competitive compiler, ultimately.

Hopefully the good start he's provided will attract the attention of someone with time and knowledge enough to carry it through to the finish line.  Or perhaps TLO himself will find some way to make it happen.

[Dreaming] Before too long here, not supporting 64-bit architectures is going to become an untenable position for any compiler.  As that day approaches, eventually Walter himself may decide that switching to the LLVM back-end makes the most sense.  If so then LLVMDC would give him a nice start, I think.

--bb

Bill Baxter:
> eventually Walter himself may decide that switching to the LLVM back-end makes the most sense.

LLVM may allow D to do certain things that today are difficult, like run-time creation of routines (they can be created by macros too), and other dynamic things you can see in C#3.x, etc. With that you can create something that even CommonLisp programmers may find attractive ;-)

Bye,
bearophile

bearophile wrote:
> Bill Baxter:
>> eventually Walter himself may decide that switching to the LLVM back-end makes the most sense.
> 
> LLVM may allow D to do certain things that today are difficult, like run-time creation of routines (they can be created by macros too), and other dynamic things you can see in C#3.x, etc. With that you can create something that even CommonLisp programmers may find attractive ;-)

Not to mention that it should fix a raft of other long-standing bugs that have to do with OPTLINK.  I'm pretty convinced that LLVM is the way to go long term.  It would free Walter up from having to deal with back end issues, but still allow him to tinker with the back end or contribute patches to the LLVM team if he needs something to be fixed for D.  It would allow D to benefit from a world wide community working on porting to new back-end targets, and making improvements to the optimizer etc. Not to mention allowing D to piggyback on the corporate support from the likes of Apple that is going into LLVM right now.

I see basically no down sides to such a move, other than making the move would initially be a big time suck.  But I think the writing is on the wall that OPTLINK will have to be replaced eventually one way or another.  Going with LLVM looks to be the best way to do that in terms of cost/benefit ratios.

--bb

On Thu, 14 Feb 2008 09:11:58 +0900, Bill Baxter wrote:

> bearophile wrote:
>> Bill Baxter:
>>> eventually Walter himself may decide that switching to the LLVM back-end makes the most sense.
>> 
>> LLVM may allow D to do certain things that today are difficult, like run-time creation of routines (they can be created by macros too), and other dynamic things you can see in C#3.x, etc. With that you can create something that even CommonLisp programmers may find attractive ;-)
> 
> Not to mention that it should fix a raft of other long-standing bugs that have to do with OPTLINK.  I'm pretty convinced that LLVM is the way to go long term.  It would free Walter up from having to deal with back end issues, but still allow him to tinker with the back end or contribute patches to the LLVM team if he needs something to be fixed for D.  It would allow D to benefit from a world wide community working on porting to new back-end targets, and making improvements to the optimizer etc. Not to mention allowing D to piggyback on the corporate support from the likes of Apple that is going into LLVM right now.
> 
> I see basically no down sides to such a move, other than making the move would initially be a big time suck.  But I think the writing is on the wall that OPTLINK will have to be replaced eventually one way or another.  Going with LLVM looks to be the best way to do that in terms of cost/benefit ratios.
> 
> --bb

Maybe it could be something to push for, for D3.

On Wed, 13 Feb 2008 14:13:48 +0900, Bill Baxter wrote:

> Graham St Jack wrote:
>>> I've got hopes that one day an LLVM-based D compiler will be the one D
>>> compiler to rule them all.
>>> (There's a start of one limping along at
>>> http://www.dsource.org/projects/llvmdc).
>> 
>> Sounds great - looking forward to it.
> 
> Yeh, me too.  But I'm afraid that unless it becomes someone's full-time obsession, realistically it's never going to be ready for prime time.
> 
> Thomas L. Olsen has gotten it off to a good start as his hobby project, despite apparently starting from zero knowledge of how LLVM works.  But I don't think his spare time effort is really going to be enough to create a competitive compiler, ultimately.
> 
> Hopefully the good start he's provided will attract the attention of someone with time and knowledge enough to carry it through to the finish line.  Or perhaps TLO himself will find some way to make it happen.
> 
> [Dreaming] Before too long here, not supporting 64-bit architectures is going to become an untenable position for any compiler.  As that day approaches, eventually Walter himself may decide that switching to the LLVM back-end makes the most sense.  If so then LLVMDC would give him a nice start, I think.

I agree.