Shixiong Xu via llvm-dev
2016-Aug-17 10:48 UTC
[llvm-dev] Cost model is missing in InstCombiner
Hi,
I think canEvaluateTruncated() in InstCombiner needs use cost model to decide
whether perform optimization or not. Without cost model from
TargetTransformInfo, aggressively optimizing IR in vector types according to the
number of bits demanded may lead to scalarization of vector operations. For
example, if the input IR is:
%wide.load25 = load <32 x i8>, <32 x i8>* %231, align 1
%232 = zext <32 x i8> %wide.load25 to <32 x i16>
%233 = mul nuw nsw <32 x i16> %232, %164
...
%237 = trunc <32 x i16> %233 to <32 x i8>
store <32 x i8> %237, <32 x i8>* %236, align 1
ICE: EvaluateInDifferentType converting expression type to avoid cast: %9 =
trunc <32 x i16> %6 to <32 x i8>
IC: ADD: %6 = mul <32 x i8> %wide.load25, %wide.load
IC: Replacing %10 = trunc <32 x i16> %7 to <32 x i8>
with %6 = mul <32 x i8> %wide.load25, %wide.load
If the target doesn't have support for mul <32 x i8>, the inst
combiner will yield less profitable code.
Cheers,
Shixiong (Jason) Xu
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20160817/961996b5/attachment.html>
Mehdi Amini via llvm-dev
2016-Aug-18 16:05 UTC
[llvm-dev] Cost model is missing in InstCombiner
+David M.> On Aug 17, 2016, at 3:48 AM, Shixiong Xu via llvm-dev <llvm-dev at lists.llvm.org> wrote: > > Hi, > > I think canEvaluateTruncated() in InstCombiner needs use cost model to decide whether perform optimization or not.I’ve always seen InstCombine as doing “canonicalization” of the IR and not “optimization”. So the output of InstCombine should be in a form that is the most suitable for further analyses and transformations. I’m not sure how this view fits with TTI, I may not be incompatible if used within some limits I guess? — Mehdi> Without cost model from TargetTransformInfo, aggressively optimizing IR in vector types according to the number of bits demanded may lead to scalarization of vector operations. For example, if the input IR is: > > %wide.load25 = load <32 x i8>, <32 x i8>* %231, align 1 > %232 = zext <32 x i8> %wide.load25 to <32 x i16> > %233 = mul nuw nsw <32 x i16> %232, %164 > … > %237 = trunc <32 x i16> %233 to <32 x i8> > store <32 x i8> %237, <32 x i8>* %236, align 1 > > ICE: EvaluateInDifferentType converting expression type to avoid cast: %9 = trunc <32 x i16> %6 to <32 x i8> > IC: ADD: %6 = mul <32 x i8> %wide.load25, %wide.load > IC: Replacing %10 = trunc <32 x i16> %7 to <32 x i8> > with %6 = mul <32 x i8> %wide.load25, %wide.load > > If the target doesn’t have support for mul <32 x i8>, the inst combiner will yield less profitable code. > > Cheers, > > Shixiong (Jason) Xu > > _______________________________________________ > LLVM Developers mailing list > llvm-dev at lists.llvm.org <mailto:llvm-dev at lists.llvm.org> > http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev <http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev>-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20160818/c047b139/attachment.html>
Hal Finkel via llvm-dev
2016-Aug-18 16:10 UTC
[llvm-dev] Cost model is missing in InstCombiner
----- Original Message -----> From: "Mehdi Amini via llvm-dev" <llvm-dev at lists.llvm.org> > To: "Shixiong Xu" <shixiong at cadence.com> > Cc: llvm-dev at lists.llvm.org > Sent: Thursday, August 18, 2016 11:05:35 AM > Subject: Re: [llvm-dev] Cost model is missing in InstCombiner> +David M.> > On Aug 17, 2016, at 3:48 AM, Shixiong Xu via llvm-dev < > > llvm-dev at lists.llvm.org > wrote: >> > Hi, >> > I think canEvaluateTruncated() in InstCombiner needs use cost model > > to decide whether perform optimization or not. > > I’ve always seen InstCombine as doing “canonicalization” of the IR > and not “optimization”. So the output of InstCombine should be in a > form that is the most suitable for further analyses and > transformations.This is exactly our traditional view. Why can the backend not be fixed to generate better code for mul <32 x i8>? It looks like the widening in the IR is something natural to get from legalization (if you set up the correct promotion preferences in *ISelLowering). -Hal> I’m not sure how this view fits with TTI, I may not be incompatible > if used within some limits I guess?> — > Mehdi> > Without cost model from TargetTransformInfo, aggressively > > optimizing > > IR in vector types according to the number of bits demanded may > > lead > > to scalarization of vector operations. For example, if the input IR > > is: >> > %wide.load25 = load <32 x i8>, <32 x i8>* %231, align 1 > > > %232 = zext <32 x i8> %wide.load25 to <32 x i16> > > > %233 = mul nuw nsw <32 x i16> %232, %164 > > > … > > > %237 = trunc <32 x i16> %233 to <32 x i8> > > > store <32 x i8> %237, <32 x i8>* %236, align 1 >> > ICE: EvaluateInDifferentType converting expression type to avoid > > cast: %9 = trunc <32 x i16> %6 to <32 x i8> > > > IC: ADD: %6 = mul <32 x i8> %wide.load25, %wide.load > > > IC: Replacing %10 = trunc <32 x i16> %7 to <32 x i8> > > > with %6 = mul <32 x i8> %wide.load25, %wide.load >> > If the target doesn’t have support for mul <32 x i8>, the inst > > combiner will yield less profitable code. >> > Cheers, >> > Shixiong (Jason) Xu > > > _______________________________________________ > > > LLVM Developers mailing list > > > llvm-dev at lists.llvm.org > > > http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev > > _______________________________________________ > LLVM Developers mailing list > llvm-dev at lists.llvm.org > http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev-- Hal Finkel Assistant Computational Scientist Leadership Computing Facility Argonne National Laboratory -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20160818/68cc311c/attachment-0001.html>