[LLVMdev] [PATCH] parallel loop awareness to the LoopVectorizer

[Renato Golin]

On 29 January 2013 10:03, Pekka Jääskeläinen <pekka.jaaskelainen at tut.fi>wrote:

> Sounds reasonable, as it's too intrusive to start making all the passes
> parallel loop-aware.
>

+1. It will bloat IR a bit, but we can turn it on only when vectorization
is enabled, when it's likely that people won't be looking at IR anyway.


In the restricted ptr case one wants to mark pointers with some kind of
> context marker (e.g. a function name + pointer identification) which
> communicates that the accesses through that pointer do not alias with
> other pointers in the same function context.
>

You may have issues with inlining, ie. have to teach it to change the
metadata to cope with the function changing, since inlining can expose many
forms of parallelism.

cheers,
--renato
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20130129/46485141/attachment.html>