Speculative prefetching: report

[Zdenek Dvorak]

Hello,

> Before pushing it to mainline I would like see the
> quadratic bottlenect of bsi_for_stmt resolved.  I don't know where the
> discussion surrounding Zdenek's patch went but I guess without that we
> can fix it by keeping around BSIs instead of direct statement pointers
> in the interface that is straightforward change.

this does not work (or at least it is likely to cause problems), since
BSI's may get invalidated when basic blocks are split.

Zdenek