[HN Gopher] High performance array programming in Petalisp
       ___________________________________________________________________
        
       High performance array programming in Petalisp
        
       Author : medo-bear
       Score  : 51 points
       Date   : 2024-05-06 21:41 UTC (1 days ago)
        
 (HTM) web link (zenodo.org)
 (TXT) w3m dump (zenodo.org)
        
       | guenthert wrote:
       | "Not only is this JIT-compiler just 8 lines of code, it is also
       | 20 times faster than invoking GCC or Clang on a roughly
       | equivalent piece of C code."
       | 
       | So about as fast as TinyCcompiler (tcc)? Which produces faster
       | code (in the context of HPC, elegance and ease of development
       | takes a backseat to sheer performance, methinks)?
        
         | medo-bear wrote:
         | Why would you use C, or even assembly, when machine code is
         | faster
        
       | guenthert wrote:
       | iiuc, that JIT-compiler uses the CL-image's compile function.
       | That won't do any good, if in a HPC environment the actual heavy
       | lifting is to be done on equipment other than the main CPU, e.g.
       | a GPGPU or (earlier) Xeon Phi.
        
         | medo-bear wrote:
         | In the paper the author talks about compiling to CUDA
        
           | guenthert wrote:
           | "We already have code generators for turning kernels into
           | Com- mon Lisp code and for turning a subset of kernels into
           | C++ or CUDA code. Right now, our strategy is to use C++ and
           | GCC when possible, and Common Lisp code otherwise. In the
           | future, we plan to make the C++ generator obsolete by using
           | SIMD optimized Common Lisp instead. Doing so would build on
           | our previous work on s b - s i m d that we presented at the
           | 15th European Lisp Symposium in Porto"
           | 
           | My take from that is, that for CUDA external tools are used
           | today. Still, interesting work.
        
             | medo-bear wrote:
             | What do you mean by external tools?
        
       | dang wrote:
       | Related (but tiny):
       | 
       |  _Petalisp: A DSL in CL intended for high performance array-based
       | computing_ - https://news.ycombinator.com/item?id=17170393 - May
       | 2018 (1 comment)
        
       | hr0m wrote:
       | Wait, I know the author and will share the link so he can join
       | here.
        
       ___________________________________________________________________
       (page generated 2024-05-07 23:01 UTC)