Are you also planning a LLVM backend for ghc, in a general sense, or just for the accelerated work you're doing? It seems to me that ghc itself could be well served with a LLVM backend, especially if one relies on the JIT mode. That could help identify code paths in the core and runtime that are infrequently used, further optimizing ghc's overall performance.<br>
<br>That's the gist behind Latter's work to accelerate the OpenGL stack on Mac OS. The LLVM JIT determines which code paths are actually taken, generating code for only those paths.<br><br><br>-scooter<br><br>PS: I do stand behind my assertion regarding the Cell. I'm the CellSPU backend hacker for LLVM and I've pretty much stopped work on it because IBM has more or less killed the chip.<br>
<br><div class="gmail_quote">On Thu, Feb 4, 2010 at 3:07 PM, Manuel M T Chakravarty <span dir="ltr"><<a href="mailto:chak@cse.unsw.edu.au">chak@cse.unsw.edu.au</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Felipe Lessa:<br>
>> I would suggest that any GSoC project in this space should be based<br>
>> on D.A.Accelerate (rather than DPH), simply because the code base is<br>
>> much smaller and more accessible. There is not much point in<br>
>> writing a CUDA backend, as we already have a partially working one<br>
>> that we are going to release in due course. However, I repeatedly<br>
>> had people asking for an OpenCL backend. So, there appears to be<br>
>> some demand for that (and it's the right thing to do, given that<br>
>> CUDA is tied to a single company). An OpenCL backend for<br>
>> D.A.Accelerate also appears to be in the scope of what a good coder<br>
>> can achieve in the timeframe of a GSoC project. (To be precise, I<br>
>> think, a good coder can implement a working backend in that<br>
>> timeframe, but it will probably require more work to generate well<br>
>> optimised code.)<br>
><br>
> Thanks, that's very interesting. What about an LLVM backend, would it<br>
> be useful? Perhaps it would be possible to use its vector operations<br>
> to use SIMD instructions of modern CPUs (I think GHC isn't there yet,<br>
> right?). This is just a thought :).<br>
<br>
I'm currently implementing an LLVM backend. I'm not planning on using SIMD instructions in the first version, but it is an interesting idea for when a basic LLVM works.<br>
<font color="#888888"><br>
Manuel<br>
</font><div><div></div><div class="h5"><br>
_______________________________________________<br>
Glasgow-haskell-users mailing list<br>
<a href="mailto:Glasgow-haskell-users@haskell.org">Glasgow-haskell-users@haskell.org</a><br>
<a href="http://www.haskell.org/mailman/listinfo/glasgow-haskell-users" target="_blank">http://www.haskell.org/mailman/listinfo/glasgow-haskell-users</a><br>
</div></div></blockquote></div><br>