[PATCH] Remove 4,096 loop scale limitation.

[Xinliang David Li]

On Tue, Mar 31, 2015 at 6:35 AM, Chandler Carruth <chandlerc at gmail.com>
wrote:

> Generally looks good, seems like a nice practical step. Should let Duncan
> check the math before committing.
>
>
> ================
> Comment at: lib/Analysis/BlockFrequencyInfoImpl.cpp:339-348
> @@ -344,1 +338,12 @@
>
> +  // Infinite loops need special handling. If we give the back edge an
> infinite
> +  // mass, they may saturate all the other scales in the function down to
> 1,
> +  // making all the other region temperatures look exactly the same.
> Choose an
> +  // arbitrary scale to avoid these issues.
> +  //
> +  // TODO: An alternate way would be to select a symbolic scale which is
> later
> +  // replaced to be the maximum of all computed scales plus 1. This would
> +  // appropriately describe the loop as having a large scale, without
> skewing
> +  // the final frequency computation.
> +  const Scaled64 InifiniteLoopScale(1, 12);
> +
> ----------------
> Mostly to check that I'm understanding this correctly, this will still
> result in really bizarre behavior where in some cases a non-infinite outer
> loop will be scaled higher than a finite inner loop.
>

you mean 'higher than infinite inner loop' here?

No it does not matter as we are approximating here.  Static 'infinite' loop
does not mean it will have high dynamic iteration count or vise versa.

David




>
> Anyways, I would make this a FIXME as this may end up mattering more than
> I would like in practice.
>
> ================
> Comment at: lib/Analysis/BlockFrequencyInfoImpl.cpp:430-431
> @@ -429,4 +432,1 @@
>    // room to differentiate small, unequal numbers.
> -  //
> -  // TODO: fix issues downstream so that ScalingFactor can be
> -  // Scaled64(1,64)/Max.
> ----------------
> I would probably keep a FIXME here that we should be able to use the full
> 64-bits... Limiting this to 32-bits seems harmless but wasteful.
>
> ================
> Comment at: test/Analysis/BlockFrequencyInfo/bad_input.ll:35
> @@ -34,2 +34,3 @@
>
> -; Check that the loop scale maxes out at 4096, giving 2048 here.
> +; Check that the infinite loop is arbitrarily scaled to max out a 4096,
> +; giving 2048 here.
> ----------------
> "scaled to max out at" -- missing 't'.
>
> http://reviews.llvm.org/D8718
>
> EMAIL PREFERENCES
>   http://reviews.llvm.org/settings/panel/emailpreferences/
>
>
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20150331/7fcf6b4c/attachment.html>