The Aparapi project was open sourced today take a look at
http://aparapi.googlecode.com. Although previous comments had listed
concerns with Aparapi code restrictions, it would be great to work
with developers of languages like Clojure to see what features of
Aparapi might be useful.
Gary
On S
> In my opinion It would be wiser to target OpenCL first. So this effort
> will not depend on particular hardware vendor.
Sadly, I've found yet another limitation of OpenCL over the weekend.
Apparently OpenCL doesn't support GPU function pointers. Without
function pointers, and with OpenCL only su
On Fri, Sep 9, 2011 at 1:43 AM, Timothy Baldridge wrote:
> I've been kicking around an idea for some time, of starting a
> Clojure->CUDA compiler. I would like to start a discussion about this
> to figure out what some possible solutions are. First of all let me
> start a simple fact list:
>
> CUD
C in S-expression approach might be helpful, if you just want to write
C in Clojure syntax.
The advantage of this pure translator approach is you can fully
recognize what you do.
Such implementation already exist in Scheme.
http://practical-scheme.net/gauche/man/gauche-refe_76.html
I think transla
I didn't see Penumbra or Calx mentioned in this thread yet (sorry if they
were and I missed it).
See this thread:
http://groups.google.com/group/clojure/browse_thread/thread/5703f75f2ccf0bec
Penumbra:
https://github.com/ztellman/penumbra
Calx:
https://github.com/ztellman/calx
Andy
On Fri,
> If it is not,
> then looking at Aparapi might be the better choice, and the AMD
> developers have openly stated that they would like to work with
> someone from the Clojure community on getting their openCL api working
> with Clojure.
Sadly, Aparapi is very, very limited. Here is an example:
ht
If this is true, then CUDA may be the better choice. If it is not,
then looking at Aparapi might be the better choice, and the AMD
developers have openly stated that they would like to work with
someone from the Clojure community on getting their openCL api working
with Clojure.
http://groups.goo
> Dissecting PyCuda might give you a headstart :)
from everything I can tell, jcuda
(http://www.jcuda.de/jcuda/JCuda.html) already contains everything
PyCUDA does. That is, both require you to write your GPU kernels in
C++. This is what I'm trying to avoid with CUDA Clojure. I want to
write my GPU
Dissecting PyCuda might give you a headstart :)
On 9 September 2011 06:05, Michael Jaaka wrote:
> Why not to translate to any lang? I though about translating clojure
> to php source code. It differs from jvm and crl (.net) approach that
> it is not translated to uni language to be run on one pl
Why not to translate to any lang? I though about translating clojure
to php source code. It differs from jvm and crl (.net) approach that
it is not translated to uni language to be run on one platform but
translated to form/source understood by targeted hosting platform and
ran on it.
The clojures
> Why not opencl?
Yes if we were going for a very simple OpenCL in Clojure Syntax, the
yes OpenCL is usable. However, last I checked, the OpenCL kernel
language did not support virtual functions or dynamic memory
allocation. That is only the CPU is allowed to allocate memory. This
is something tha
I haven't looked at aparapi [1] in detail, but it seems interesting
and potentially useful for your quest:
[1] http://developer.amd.com/zones/java/aparapi/pages/default.aspx
-Jason
On Sep 8, 11:43 am, Timothy Baldridge wrote:
> I've been kicking around an idea for some time, of starting a
> Clo
Why not opencl? This way you would be able to run the code in any
hardware and even a hybrid approach cpu+gpu(AMD and NVIDIA), you could
even think about webcl (javascript to access the gpu) which I think
has some form of GC. Besides, opencl code is very similar to CUDA
code.
Whatever you do, thi
First, I will agree with your self diagnosis of crazy.
Second, I think this will be very interesting. I am personally
interested in CUDA and have written a fair amount of CUDA stuff to see
how it works. My initial reaction is that I would hate to have a tool
that removes any of the functiona
I've been kicking around an idea for some time, of starting a
Clojure->CUDA compiler. I would like to start a discussion about this
to figure out what some possible solutions are. First of all let me
start a simple fact list:
CUDA (for those who don't know) is NVIDIA's technology for writing
gener
15 matches
Mail list logo
