Stmt deep copy

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Stmt deep copy

Nathan Lawrence
Hi,

I'm trying to do loop unrolling inside of clang as part of a source to
source transformation, and I'm wondering if there is any way to deep
copy statements.

Thanks,
-- Nathan
_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev
Reply | Threaded
Open this post in threaded view
|

Re: Stmt deep copy

A.S. Bradbury
On 19 March 2010 00:45, Nathan Lawrence <[hidden email]> wrote:
> Hi,
>
> I'm trying to do loop unrolling inside of clang as part of a source to
> source transformation, and I'm wondering if there is any way to deep
> copy statements.

I've started to work on some source to source transformations in clang
and have found Olaf Krzikalla's ASTProcessing library useful. He
posted it to the mailing list here:
http://article.gmane.org/gmane.comp.compilers.clang.devel/5323/

Google told me the current development repository is available here:
https://gforge.zih.tu-dresden.de/projects/hicfd/

I must have missed it if the same functionality is already available
in clang core, but Olaf's implementation is in
Scout/clangAddons/lib/ASTProcessing/StmtClone.cpp

Alex
_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev
Reply | Threaded
Open this post in threaded view
|

Re: Stmt deep copy

Olaf Krzikalla
In reply to this post by Nathan Lawrence
Nathan Lawrence schrieb:
> I'm trying to do loop unrolling inside of clang as part of a source to
> source transformation, and I'm wondering if there is any way to deep
> copy statements.
>  
It looks as if several people work on similiar things independently.
Alex already posted all the useful links to my work. Of course, if there
are any questions feel free to ask, contribute or the like.
At the moment I'm writing a paper to promote my actual project
(source-to-source SIMD vectorization) a little bit more.
Maybe then there is a greater chance to make some of the transformation
framework a part of clang (read: I have some more time). At least I want
to get rid of the still necessary patch.
IMHO AST processing pops up rather frequently on the list lately.

Best
Olaf Krzikalla

_______________________________________________
cfe-dev mailing list
[hidden email]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev