This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Loop optimizer issues


>>>>> Diego Novillo writes:

Diego> To avoid duplicate effort, let's just raise our hand whenever we are
Diego> going to get started with something.  The initial list that I had in
Diego> mind (at the tree level) includes:

Diego> Induction variable detection
Diego> Array dependency information
Diego> Unswitching
Diego> Peeling
Diego> Interchange
Diego> Fusion
Diego> Fission

Diego> I volunteer to collect suggestions and keep them in the TODO section of
Diego> the project page together with whoever is working on it.  If there are
Diego> two proposals for the same analysis/optimization but with different
Diego> approaches, we may want to implement both and then decide which one we
Diego> like best.

	The partial list of loop optimizations from a March 2002 public
presentation about XLC compiler optimizations includes:

normalization
scalarization
fusion
nest partitioning
interchange
unroll and jam

I will see if I can provided a more complete and ordered list of TPO
optimizations. 

	Dan suggested that GCC may get some scalarization implicitly in
its other SSA optimizations.

David


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]