llvm dev - Jan 2013 - [LLVMdev] Does loop vectorizer inquire about target's SIMD capabilities?

On 3 January 2013 22:09, Nadav Rotem <nrotem at apple.com> wrote:
> The loop vectorizer is now enabled by default.
>
I thought that was just a temporary arrangement to get the feel for it, not
to actually have it on all the time (next release). Is it just for -O3 or
lower too?

This can cause problems, for instance on ARMv7, the default is that NEON is
present, but Tegra2 doesn't have NEON, only VFP. It means an optimizing
compilation that used to work on it will fail, unless you specify no NEON
or no vectorization explicitly, no?

I'd prefer if -vectorization had to be passed explicitly to get it done on
production releases of LLVM or something else (-O4 or -OV, for example).

cheers,
--renato
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20130103/c4f5e285/attachment.html>

On Jan 3, 2013, at 3:04 PM, Renato Golin <renato.golin at linaro.org>
wrote:
> On 3 January 2013 22:09, Nadav Rotem <nrotem at apple.com> wrote:
> The loop vectorizer is now enabled by default. 
> 
> I thought that was just a temporary arrangement to get the feel for it, not
to actually have it on all the time (next release). Is it just for -O3 or lower
too?
The plan it to enable it by default for the 3.3 release. Until that point we can
enable and disable it depending on the situation. It is a good idea to enable it
as soon as possible in order to catch performance regressions as soon as
possible. The vectorizer is currently enabled by default on O2 and O3, and it
runs with reduced functionality on Os.
> 
> This can cause problems, for instance on ARMv7, the default is that NEON is
present, but Tegra2 doesn't have NEON, only VFP. It means an optimizing
compilation that used to work on it will fail, unless you specify no NEON or no
vectorization explicitly, no?
At the moment ARM does not have a good cost model. It relies on the default
implementation which uses the TargetLoweing information to collect information
about the target. In theory things should just work, because TLI should know
about the vector situation.

I plan to start working on the ARM cost model early next week, and I hope that
the ARM folks will help me with this.
> 
> I'd prefer if -vectorization had to be passed explicitly to get it done
on production releases of LLVM or something else (-O4 or -OV, for example).
Clang has the '-fvectorize' and '-fno-vectorize'.  
> 
> cheers,
> --renato
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20130103/06da9d40/attachment.html>

On 3 January 2013 23:24, Nadav Rotem <nrotem at apple.com> wrote:
> I plan to start working on the ARM cost model early next week, and I hope
> that the ARM folks will help me with this.
>
This is great news! I hope to be able to help you with this. Let me know
when you start and we can divide the work.

cheers,
--renato
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20130104/2ccf9b40/attachment.html>