[PATCH] Generate warning when explicitly specified vectorization fails

Tyler Nowicki tnowicki at apple.com
Wed Jul 9 19:40:38 PDT 2014

If these pragmas are silently ignored when then they don’t have much value. A big part of their value come from the confirmation that they are working. The warning will only be generated on loops with #pragma clang loop vectorize/interleave/unroll specified. It can also be silenced using the command line option -Wno-optimization-warning.

I’m still looking into the clang in 32-bit mode question posed earlier. Are there any other comments on the patches?



On Jul 7, 2014, at 10:45 AM, Mark Heffernan <meheff at google.com> wrote:

> On Thu, Jul 3, 2014 at 9:23 AM, Diego Novillo <dnovillo at google.com> wrote:
> This change is raising the severity of this diagnostic. Do we really want to cause build failures in this situation when the code is built with -Werror?
> I think I would be fine with it, if this diagnostic was only triggered when the user specified a #pragma, but if the failure is in the vectorizer, with no hints from the user, then I think a warning is too strong. Even when using #pragma, I'm not sure I'd like a warning here.
> I'd put my vote in for having a warning if a #pragma optimization directive could not be followed.  This assumes the warning is instructive enough that the user can easily fix the issue.  I added support for loop unrolling pragmas (following Tyler's vectorization work) so the focus for us is loop unrolling.  For the target we care about (CUDA/GPU) a missed loop unroll can be a massive performance loss and the user should know about it.  Whatever is decided, I'll make the diagnostic severity of loop unrolling pragma diagnostics match that of the vectorization ones.
> Mark
>        return false;
>      }
> @@ -1276,8 +1292,7 @@
>        emitOptimizationRemarkAnalysis(
>            F->getContext(), DEBUG_TYPE, *F, L->getStartLoc(),
>            "loop not vectorized due to NoImplicitFloat attribute");
> -      emitOptimizationRemarkMissed(F->getContext(), DEBUG_TYPE, *F,
> -                                   L->getStartLoc(), Hints.emitRemark());
> +      emitMissedWarning(F, L, Hints);
> Likewise.
> Diego.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20140709/bc096bbe/attachment.html>

More information about the cfe-commits mailing list