Parallel Compiler Representation

Xinan TANG <tang@binkley.cs.mcgill.ca>
Tue, 8 Mar 1994 01:34:32 GMT

          From comp.compilers
Related articles
Parallel Compiler Representation tang@binkley.cs.mcgill.ca (Xinan TANG) (1994-03-08)
Re: Parallel Compiler Representation havlak@cs.umd.edu (1994-03-21)
| List of all articles for this month |
Newsgroups: comp.compilers
From: Xinan TANG <tang@binkley.cs.mcgill.ca>
Keywords: optimize, parallel, question
Organization: Compilers Central
Date: Tue, 8 Mar 1994 01:34:32 GMT

Hi


    I have a question concerning on intermediate representation of
Parallel compilers.


    There are many forms for compiler optimizations such as the CFG, SSA,
PDG, PDW etc. but all of them are not good enough to represent data
dependency for arrays and pointer introduced recursive data structures.
Actually, the parallelism just comes from array based loop and disjoint
recursive function calls.


    When we are talking parallelism from array based loops, the techniques
used are subscript base dependence analysis, loop transformation and
parallelization. What's the place of internal representation? Isn't it
as important as in the compiler optimization? Can we say no matter
what's internal representation if you represent loop as "normal form"
that's OK? Can anyone who are writing sort of parallel compiler give me
a hint on this point.


Thank you


Tang
--

Post a followup to this message

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