Re: [tree-ssa] Re: auto vectorization in gcc

[Pop Sébastian <pop at gauvain dot u-strasbg dot fr>]

On Wed, Jul 23, 2003 at 07:50:40PM +0300, Dorit Naishlos wrote:
> 
> > This is very much along the same lines as I was envisaging.  Given we
> > currently have some time and resources that could be used on a project
> > like this would you be interested / able tocolaborate on this project?
> 
> Sure! (there's a lot to be done...)
> 
> We're still at the design phase - trying to decide which
> vectorization scheme we want to implement, what is the minimal
> essential infrastructure required in order to vectorize even
> the simplest loop/code-sequence, and how much of this infrastructure
> is already available, or is being developed and will be available (and
> when).
> 
> I understand that Sebastian Pop is working on IV detection,
Yes, I'm on this project and intends to post a version of my IV analyzer 
by the end of this month.  This is still an experimental analyzer 
that transforms the integer expressions from loops into evolution envelopes,
from which the dependence relations/distance vectors are then extracted.

  http://icps.u-strasbg.fr/~pop/DEA_03_Pop.pdf
  http://icps.u-strasbg.fr/~pop/DEA_03_Pop.ps.gz

> and that data dependence analysis will be based on that infrastructure.
> Where does that stand? Sebastian, were you planning to work on the
> dependence graph too?
Yes, this consists in implementing an extraction function from the 
envelope chains of recurrences.  

I know that Zdenek Dvorak works on the RTL level IV detection that could
be less experimental than my analysis.  It is maybe possible to adapt 
his analyzer to work as well on trees.  I have not investigated this solution,
but that could factor down the number of new lines of code for having IV 
detection at tree level.

> 
> The project's web page states that loop normalization
> and loop canonicalization are on the todo list. Is someone working
> on that? (Daniel Berlin?) I guess that's pending the IV detection work too?
> 
> Jan Hubicka wrote (http://gcc.gnu.org/ml/gcc/2003-07/msg00185.html):
> > "...I am not sure whether Tree-SSA is quite
> > ready for playing with things like, say, loop unrolling when there are
> > still no ways to duplicate basic blocks and redirect edges..."
> 
> I didn't see loop unrolling on the todo list though. I saw that there's
> an intention to borrow the loop unroller from the rtlopt branch.
> Where does that stand?
> 
We need the number of iterations in a loop for this optimization, this also 
waits for the IV analysis.


Sebastian