Compiler optimizations on GPU?

"Inderaj Bains" <inderaj@gmail.com>
Wed, 30 Apr 2008 13:53:02 -0700

          From comp.compilers

Related articles
Compiler optimizations on GPU? shreyas76@gmail.com (shrey) (2008-04-29)
Re: Compiler optimizations on GPU? thisisgiri@gmail.com (Giridhar S) (2008-04-29)
Re: Compiler optimizations on GPU? torbenm@app-2.diku.dk (2008-04-30)
Compiler optimizations on GPU? inderaj@gmail.com (Inderaj Bains) (2008-04-30)
Re: Compiler optimizations on GPU? al407@cam.ac.uk (Anton Lokhmotov) (2008-05-06)
| List of all articles for this month |

From: "Inderaj Bains" <inderaj@gmail.com>
Newsgroups: comp.compilers
Date: Wed, 30 Apr 2008 13:53:02 -0700
Organization: Compilers Central
References: 08-04-109
Keywords: optimize
Posted-Date: 03 May 2008 17:43:54 EDT

It depends on the programming language you are using If you are using
a language like cuda, the parallelism is expressed well and you need
no rocket science in compilation but basic good codegen


Reducing operation counts is important on GPU's as well So most of the
same high level optimizations are required


One biggest difference is low or no cache, due to which you need to be
especially mindful of not creating too many extra lifetimes.In that
sense is tradeoff point of computation vs lifetime is different from
regular processors


Inderaj


Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.